Desktop supercomputer: what can it do?

Science.gov (United States)

Bogdanov, A.; Degtyarev, A.; Korkhov, V.

2017-12-01

The paper addresses the issues of solving complex problems that require using supercomputers or multiprocessor clusters available for most researchers nowadays. Efficient distribution of high performance computing resources according to actual application needs has been a major research topic since high-performance computing (HPC) technologies became widely introduced. At the same time, comfortable and transparent access to these resources was a key user requirement. In this paper we discuss approaches to build a virtual private supercomputer available at user's desktop: a virtual computing environment tailored specifically for a target user with a particular target application. We describe and evaluate possibilities to create the virtual supercomputer based on light-weight virtualization technologies, and analyze the efficiency of our approach compared to traditional methods of HPC resource management.

Adaptability of supercomputers to nuclear computations

International Nuclear Information System (INIS)

Asai, Kiyoshi; Ishiguro, Misako; Matsuura, Toshihiko.

1983-01-01

Recently in the field of scientific and technical calculation, the usefulness of supercomputers represented by CRAY-1 has been recognized, and they are utilized in various countries. The rapid computation of supercomputers is based on the function of vector computation. The authors investigated the adaptability to vector computation of about 40 typical atomic energy codes for the past six years. Based on the results of investigation, the adaptability of the function of vector computation that supercomputers have to atomic energy codes, the problem regarding the utilization and the future prospect are explained. The adaptability of individual calculation codes to vector computation is largely dependent on the algorithm and program structure used for the codes. The change to high speed by pipeline vector system, the investigation in the Japan Atomic Energy Research Institute and the results, and the examples of expressing the codes for atomic energy, environmental safety and nuclear fusion by vector are reported. The magnification of speed up for 40 examples was from 1.5 to 9.0. It can be said that the adaptability of supercomputers to atomic energy codes is fairly good. (Kako, I.)

Desktop supercomputer: what can it do?

International Nuclear Information System (INIS)

Bogdanov, A.; Degtyarev, A.; Korkhov, V.

2017-01-01

The paper addresses the issues of solving complex problems that require using supercomputers or multiprocessor clusters available for most researchers nowadays. Efficient distribution of high performance computing resources according to actual application needs has been a major research topic since high-performance computing (HPC) technologies became widely introduced. At the same time, comfortable and transparent access to these resources was a key user requirement. In this paper we discuss approaches to build a virtual private supercomputer available at user's desktop: a virtual computing environment tailored specifically for a target user with a particular target application. We describe and evaluate possibilities to create the virtual supercomputer based on light-weight virtualization technologies, and analyze the efficiency of our approach compared to traditional methods of HPC resource management.

TOP500 Supercomputers for November 2004

Energy Technology Data Exchange (ETDEWEB)

Strohmaier, Erich; Meuer, Hans W.; Dongarra, Jack; Simon, Horst D.

2004-11-08

24th Edition of TOP500 List of World's Fastest Supercomputers Released: DOE/IBM BlueGene/L and NASA/SGI's Columbia gain Top Positions MANNHEIM, Germany; KNOXVILLE, Tenn.; BERKELEY, Calif. In what has become a closely watched event in the world of high-performance computing, the 24th edition of the TOP500 list of the worlds fastest supercomputers was released today (November 8, 2004) at the SC2004 Conference in Pittsburgh, Pa.

TOP500 Supercomputers for June 2003

Energy Technology Data Exchange (ETDEWEB)

Strohmaier, Erich; Meuer, Hans W.; Dongarra, Jack; Simon, Horst D.

2003-06-23

21st Edition of TOP500 List of World's Fastest Supercomputers Released MANNHEIM, Germany; KNOXVILLE, Tenn.;&BERKELEY, Calif. In what has become a much-anticipated event in the world of high-performance computing, the 21st edition of the TOP500 list of the world's fastest supercomputers was released today (June 23, 2003). The Earth Simulator supercomputer built by NEC and installed last year at the Earth Simulator Center in Yokohama, Japan, with its Linpack benchmark performance of 35.86 Tflop/s (teraflops or trillions of calculations per second), retains the number one position. The number 2 position is held by the re-measured ASCI Q system at Los Alamos National Laboratory. With 13.88 Tflop/s, it is the second system ever to exceed the 10 Tflop/smark. ASCIQ was built by Hewlett-Packard and is based on the AlphaServerSC computer system.

TOP500 Supercomputers for June 2002

Energy Technology Data Exchange (ETDEWEB)

Strohmaier, Erich; Meuer, Hans W.; Dongarra, Jack; Simon, Horst D.

2002-06-20

19th Edition of TOP500 List of World's Fastest Supercomputers Released MANNHEIM, Germany; KNOXVILLE, Tenn.;&BERKELEY, Calif. In what has become a much-anticipated event in the world of high-performance computing, the 19th edition of the TOP500 list of the worlds fastest supercomputers was released today (June 20, 2002). The recently installed Earth Simulator supercomputer at the Earth Simulator Center in Yokohama, Japan, is as expected the clear new number 1. Its performance of 35.86 Tflop/s (trillions of calculations per second) running the Linpack benchmark is almost five times higher than the performance of the now No.2 IBM ASCI White system at Lawrence Livermore National Laboratory (7.2 Tflop/s). This powerful leap frogging to the top by a system so much faster than the previous top system is unparalleled in the history of the TOP500.

Status reports of supercomputing astrophysics in Japan

International Nuclear Information System (INIS)

Nakamura, Takashi; Nagasawa, Mikio

1990-01-01

The Workshop on Supercomputing Astrophysics was held at National Laboratory for High Energy Physics (KEK, Tsukuba) from August 31 to September 2, 1989. More than 40 participants of physicists, astronomers were attendant and discussed many topics in the informal atmosphere. The main purpose of this workshop was focused on the theoretical activities in computational astrophysics in Japan. It was also aimed to promote effective collaboration between the numerical experimentists working on supercomputing technique. The various subjects of the presented papers of hydrodynamics, plasma physics, gravitating systems, radiative transfer and general relativity are all stimulating. In fact, these numerical calculations become possible now in Japan owing to the power of Japanese supercomputer such as HITAC S820, Fujitsu VP400E and NEC SX-2. (J.P.N.)

Comprehensive efficiency analysis of supercomputer resource usage based on system monitoring data

Science.gov (United States)

Mamaeva, A. A.; Shaykhislamov, D. I.; Voevodin, Vad V.; Zhumatiy, S. A.

2018-03-01

One of the main problems of modern supercomputers is the low efficiency of their usage, which leads to the significant idle time of computational resources, and, in turn, to the decrease in speed of scientific research. This paper presents three approaches to study the efficiency of supercomputer resource usage based on monitoring data analysis. The first approach performs an analysis of computing resource utilization statistics, which allows to identify different typical classes of programs, to explore the structure of the supercomputer job flow and to track overall trends in the supercomputer behavior. The second approach is aimed specifically at analyzing off-the-shelf software packages and libraries installed on the supercomputer, since efficiency of their usage is becoming an increasingly important factor for the efficient functioning of the entire supercomputer. Within the third approach, abnormal jobs – jobs with abnormally inefficient behavior that differs significantly from the standard behavior of the overall supercomputer job flow – are being detected. For each approach, the results obtained in practice in the Supercomputer Center of Moscow State University are demonstrated.

Comments on the parallelization efficiency of the Sunway TaihuLight supercomputer

OpenAIRE

Végh, János

2016-01-01

In the world of supercomputers, the large number of processors requires to minimize the inefficiencies of parallelization, which appear as a sequential part of the program from the point of view of Amdahl's law. The recently suggested new figure of merit is applied to the recently presented supercomputer, and the timeline of "Top 500" supercomputers is scrutinized using the metric. It is demonstrated, that in addition to the computing performance and power consumption, the new supercomputer i...

The ETA10 supercomputer system

International Nuclear Information System (INIS)

Swanson, C.D.

1987-01-01

The ETA Systems, Inc. ETA 10 is a next-generation supercomputer featuring multiprocessing, a large hierarchical memory system, high performance input/output, and network support for both batch and interactive processing. Advanced technology used in the ETA 10 includes liquid nitrogen cooled CMOS logic with 20,000 gates per chip, a single printed circuit board for each CPU, and high density static and dynamics MOS memory chips. Software for the ETA 10 includes an underlying kernel that supports multiple user environments, a new ETA FORTRAN compiler with an advanced automatic vectorizer, a multitasking library and debugging tools. Possible developments for future supercomputers from ETA Systems are discussed. (orig.)

Integration of Panda Workload Management System with supercomputers

Science.gov (United States)

De, K.; Jha, S.; Klimentov, A.; Maeno, T.; Mashinistov, R.; Nilsson, P.; Novikov, A.; Oleynik, D.; Panitkin, S.; Poyda, A.; Read, K. F.; Ryabinkin, E.; Teslyuk, A.; Velikhov, V.; Wells, J. C.; Wenaus, T.

2016-09-01

The Large Hadron Collider (LHC), operating at the international CERN Laboratory in Geneva, Switzerland, is leading Big Data driven scientific explorations. Experiments at the LHC explore the fundamental nature of matter and the basic forces that shape our universe, and were recently credited for the discovery of a Higgs boson. ATLAS, one of the largest collaborations ever assembled in the sciences, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. The ATLAS experiment uses PanDA (Production and Data Analysis) Workload Management System for managing the workflow for all data processing on over 140 data centers. Through PanDA, ATLAS physicists see a single computing facility that enables rapid scientific breakthroughs for the experiment, even though the data centers are physically scattered all over the world. While PanDA currently uses more than 250000 cores with a peak performance of 0.3+ petaFLOPS, next LHC data taking runs will require more resources than Grid computing can possibly provide. To alleviate these challenges, LHC experiments are engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with supercomputers in United States, Europe and Russia (in particular with Titan supercomputer at Oak Ridge Leadership Computing Facility (OLCF), Supercomputer at the National Research Center "Kurchatov Institute", IT4 in Ostrava, and others). The current approach utilizes a modified PanDA pilot framework for job submission to the supercomputers batch queues and local data management, with light-weight MPI wrappers to run singlethreaded workloads in parallel on Titan's multi-core worker nodes. This implementation was tested with a variety of Monte-Carlo workloads

Applications of supercomputing and the utility industry: Calculation of power transfer capabilities

International Nuclear Information System (INIS)

Jensen, D.D.; Behling, S.R.; Betancourt, R.

1990-01-01

Numerical models and iterative simulation using supercomputers can furnish cost-effective answers to utility industry problems that are all but intractable using conventional computing equipment. An example of the use of supercomputers by the utility industry is the determination of power transfer capability limits for power transmission systems. This work has the goal of markedly reducing the run time of transient stability codes used to determine power distributions following major system disturbances. To date, run times of several hours on a conventional computer have been reduced to several minutes on state-of-the-art supercomputers, with further improvements anticipated to reduce run times to less than a minute. In spite of the potential advantages of supercomputers, few utilities have sufficient need for a dedicated in-house supercomputing capability. This problem is resolved using a supercomputer center serving a geographically distributed user base coupled via high speed communication networks

Supercomputers to transform Science

CERN Multimedia

2006-01-01

"New insights into the structure of space and time, climate modeling, and the design of novel drugs, are but a few of the many research areas that will be transforned by the installation of three supercomputers at the Unversity of Bristol." (1/2 page)

Convex unwraps its first grown-up supercomputer

Energy Technology Data Exchange (ETDEWEB)

Manuel, T.

1988-03-03

Convex Computer Corp.'s new supercomputer family is even more of an industry blockbuster than its first system. At a tenfold jump in performance, it's far from just an incremental upgrade over its first minisupercomputer, the C-1. The heart of the new family, the new C-2 processor, churning at 50 million floating-point operations/s, spawns a group of systems whose performance could pass for some fancy supercomputers-namely those of the Cray Research Inc. family. When added to the C-1, Convex's five new supercomputers create the C series, a six-member product group offering a performance range from 20 to 200 Mflops. They mark an important transition for Convex from a one-product high-tech startup to a multinational company with a wide-ranging product line. It's a tough transition but the Richardson, Texas, company seems to be doing it. The extended product line propels Convex into the upper end of the minisupercomputer class and nudges it into the low end of the big supercomputers. It positions Convex in an uncrowded segment of the market in the $500,000 to $1 million range offering 50 to 200 Mflops of performance. The company is making this move because the minisuper area, which it pioneered, quickly became crowded with new vendors, causing prices and gross margins to drop drastically.

Supercomputer debugging workshop 1991 proceedings

Energy Technology Data Exchange (ETDEWEB)

Brown, J.

1991-01-01

This report discusses the following topics on supercomputer debugging: Distributed debugging; use interface to debugging tools and standards; debugging optimized codes; debugging parallel codes; and debugger performance and interface as analysis tools. (LSP)

Supercomputer debugging workshop 1991 proceedings

Energy Technology Data Exchange (ETDEWEB)

Brown, J.

1991-12-31

This report discusses the following topics on supercomputer debugging: Distributed debugging; use interface to debugging tools and standards; debugging optimized codes; debugging parallel codes; and debugger performance and interface as analysis tools. (LSP)

The ETA systems plans for supercomputers

International Nuclear Information System (INIS)

Swanson, C.D.

1987-01-01

The ETA Systems, is a class VII supercomputer featuring multiprocessing, a large hierarchical memory system, high performance input/output, and network support for both batch and interactive processing. Advanced technology used in the ETA 10 includes liquid nitrogen cooled CMOS logic with 20,000 gates per chip, a single printed circuit board for each CPU, and high density static and dynamic MOS memory chips. Software for the ETA 10 includes an underlying kernel that supports multiple user environments, a new ETA FORTRAN compiler with an advanced automatic vectorizer, a multitasking library and debugging tools. Possible developments for future supercomputers from ETA Systems are discussed

Automatic discovery of the communication network topology for building a supercomputer model

Science.gov (United States)

Sobolev, Sergey; Stefanov, Konstantin; Voevodin, Vadim

2016-10-01

The Research Computing Center of Lomonosov Moscow State University is developing the Octotron software suite for automatic monitoring and mitigation of emergency situations in supercomputers so as to maximize hardware reliability. The suite is based on a software model of the supercomputer. The model uses a graph to describe the computing system components and their interconnections. One of the most complex components of a supercomputer that needs to be included in the model is its communication network. This work describes the proposed approach for automatically discovering the Ethernet communication network topology in a supercomputer and its description in terms of the Octotron model. This suite automatically detects computing nodes and switches, collects information about them and identifies their interconnections. The application of this approach is demonstrated on the "Lomonosov" and "Lomonosov-2" supercomputers.

PNNL supercomputer to become largest computing resource on the Grid

CERN Multimedia

2002-01-01

Hewlett Packard announced that the US DOE Pacific Northwest National Laboratory will connect a 9.3-teraflop HP supercomputer to the DOE Science Grid. This will be the largest supercomputer attached to a computer grid anywhere in the world (1 page).

Dust modelling and forecasting in the Barcelona Supercomputing Center: Activities and developments

Energy Technology Data Exchange (ETDEWEB)

Perez, C; Baldasano, J M; Jimenez-Guerrero, P; Jorba, O; Haustein, K; Basart, S [Earth Sciences Department. Barcelona Supercomputing Center. Barcelona (Spain); Cuevas, E [Izanaa Atmospheric Research Center. Agencia Estatal de Meteorologia, Tenerife (Spain); Nickovic, S [Atmospheric Research and Environment Branch, World Meteorological Organization, Geneva (Switzerland)], E-mail: carlos.perez@bsc.es

2009-03-01

The Barcelona Supercomputing Center (BSC) is the National Supercomputer Facility in Spain, hosting MareNostrum, one of the most powerful Supercomputers in Europe. The Earth Sciences Department of BSC operates daily regional dust and air quality forecasts and conducts intensive modelling research for short-term operational prediction. This contribution summarizes the latest developments and current activities in the field of sand and dust storm modelling and forecasting.

Dust modelling and forecasting in the Barcelona Supercomputing Center: Activities and developments

International Nuclear Information System (INIS)

Perez, C; Baldasano, J M; Jimenez-Guerrero, P; Jorba, O; Haustein, K; Basart, S; Cuevas, E; Nickovic, S

2009-01-01

The Barcelona Supercomputing Center (BSC) is the National Supercomputer Facility in Spain, hosting MareNostrum, one of the most powerful Supercomputers in Europe. The Earth Sciences Department of BSC operates daily regional dust and air quality forecasts and conducts intensive modelling research for short-term operational prediction. This contribution summarizes the latest developments and current activities in the field of sand and dust storm modelling and forecasting.

Supercomputers Of The Future

Science.gov (United States)

Peterson, Victor L.; Kim, John; Holst, Terry L.; Deiwert, George S.; Cooper, David M.; Watson, Andrew B.; Bailey, F. Ron

1992-01-01

Report evaluates supercomputer needs of five key disciplines: turbulence physics, aerodynamics, aerothermodynamics, chemistry, and mathematical modeling of human vision. Predicts these fields will require computer speed greater than 10(Sup 18) floating-point operations per second (FLOP's) and memory capacity greater than 10(Sup 15) words. Also, new parallel computer architectures and new structured numerical methods will make necessary speed and capacity available.

NASA Advanced Supercomputing Facility Expansion

Science.gov (United States)

Thigpen, William W.

2017-01-01

The NASA Advanced Supercomputing (NAS) Division enables advances in high-end computing technologies and in modeling and simulation methods to tackle some of the toughest science and engineering challenges facing NASA today. The name "NAS" has long been associated with leadership and innovation throughout the high-end computing (HEC) community. We play a significant role in shaping HEC standards and paradigms, and provide leadership in the areas of large-scale InfiniBand fabrics, Lustre open-source filesystems, and hyperwall technologies. We provide an integrated high-end computing environment to accelerate NASA missions and make revolutionary advances in science. Pleiades, a petaflop-scale supercomputer, is used by scientists throughout the U.S. to support NASA missions, and is ranked among the most powerful systems in the world. One of our key focus areas is in modeling and simulation to support NASA's real-world engineering applications and make fundamental advances in modeling and simulation methods.

ATLAS Software Installation on Supercomputers

CERN Document Server

Undrus, Alexander; The ATLAS collaboration

2018-01-01

PowerPC and high performance computers (HPC) are important resources for computing in the ATLAS experiment. The future LHC data processing will require more resources than Grid computing, currently using approximately 100,000 cores at well over 100 sites, can provide. Supercomputers are extremely powerful as they use resources of hundreds of thousands CPUs joined together. However their architectures have different instruction sets. ATLAS binary software distributions for x86 chipsets do not fit these architectures, as emulation of these chipsets results in huge performance loss. This presentation describes the methodology of ATLAS software installation from source code on supercomputers. The installation procedure includes downloading the ATLAS code base as well as the source of about 50 external packages, such as ROOT and Geant4, followed by compilation, and rigorous unit and integration testing. The presentation reports the application of this procedure at Titan HPC and Summit PowerPC at Oak Ridge Computin...

JINR supercomputer of the module type for event parallel analysis

International Nuclear Information System (INIS)

Kolpakov, I.F.; Senner, A.E.; Smirnov, V.A.

1987-01-01

A model of a supercomputer with 50 million of operations per second is suggested. Its realization allows one to solve JINR data analysis problems for large spectrometers (in particular DELPHY collaboration). The suggested module supercomputer is based on 32-bit commercial available microprocessor with a processing rate of about 1 MFLOPS. The processors are combined by means of VME standard busbars. MicroVAX-11 is a host computer organizing the operation of the system. Data input and output is realized via microVAX-11 computer periphery. Users' software is based on the FORTRAN-77. The supercomputer is connected with a JINR net port and all JINR users get an access to the suggested system

Supercomputers and quantum field theory

International Nuclear Information System (INIS)

Creutz, M.

1985-01-01

A review is given of why recent simulations of lattice gauge theories have resulted in substantial demands from particle theorists for supercomputer time. These calculations have yielded first principle results on non-perturbative aspects of the strong interactions. An algorithm for simulating dynamical quark fields is discussed. 14 refs

Supercomputer applications in nuclear research

International Nuclear Information System (INIS)

Ishiguro, Misako

1992-01-01

The utilization of supercomputers in Japan Atomic Energy Research Institute is mainly reported. The fields of atomic energy research which use supercomputers frequently and the contents of their computation are outlined. What is vectorizing is simply explained, and nuclear fusion, nuclear reactor physics, the hydrothermal safety of nuclear reactors, the parallel property that the atomic energy computations of fluids and others have, the algorithm for vector treatment and the effect of speed increase by vectorizing are discussed. At present Japan Atomic Energy Research Institute uses two systems of FACOM VP 2600/10 and three systems of M-780. The contents of computation changed from criticality computation around 1970, through the analysis of LOCA after the TMI accident, to nuclear fusion research, the design of new type reactors and reactor safety assessment at present. Also the method of using computers advanced from batch processing to time sharing processing, from one-dimensional to three dimensional computation, from steady, linear to unsteady nonlinear computation, from experimental analysis to numerical simulation and so on. (K.I.)

Computational plasma physics and supercomputers

International Nuclear Information System (INIS)

Killeen, J.; McNamara, B.

1984-09-01

The Supercomputers of the 80's are introduced. They are 10 to 100 times more powerful than today's machines. The range of physics modeling in the fusion program is outlined. New machine architecture will influence particular codes, but parallel processing poses new coding difficulties. Increasing realism in simulations will require better numerics and more elaborate mathematics

Mistral Supercomputer Job History Analysis

OpenAIRE

Zasadziński, Michał; Muntés-Mulero, Victor; Solé, Marc; Ludwig, Thomas

2018-01-01

In this technical report, we show insights and results of operational data analysis from petascale supercomputer Mistral, which is ranked as 42nd most powerful in the world as of January 2018. Data sources include hardware monitoring data, job scheduler history, topology, and hardware information. We explore job state sequences, spatial distribution, and electric power patterns.

Interactive real-time nuclear plant simulations on a UNIX based supercomputer

International Nuclear Information System (INIS)

Behling, S.R.

1990-01-01

Interactive real-time nuclear plant simulations are critically important to train nuclear power plant engineers and operators. In addition, real-time simulations can be used to test the validity and timing of plant technical specifications and operational procedures. To accurately and confidently simulate a nuclear power plant transient in real-time, sufficient computer resources must be available. Since some important transients cannot be simulated using preprogrammed responses or non-physical models, commonly used simulation techniques may not be adequate. However, the power of a supercomputer allows one to accurately calculate the behavior of nuclear power plants even during very complex transients. Many of these transients can be calculated in real-time or quicker on the fastest supercomputers. The concept of running interactive real-time nuclear power plant transients on a supercomputer has been tested. This paper describes the architecture of the simulation program, the techniques used to establish real-time synchronization, and other issues related to the use of supercomputers in a new and potentially very important area. (author)

Porting Ordinary Applications to Blue Gene/Q Supercomputers

Energy Technology Data Exchange (ETDEWEB)

Maheshwari, Ketan C.; Wozniak, Justin M.; Armstrong, Timothy; Katz, Daniel S.; Binkowski, T. Andrew; Zhong, Xiaoliang; Heinonen, Olle; Karpeyev, Dmitry; Wilde, Michael

2015-08-31

Efficiently porting ordinary applications to Blue Gene/Q supercomputers is a significant challenge. Codes are often originally developed without considering advanced architectures and related tool chains. Science needs frequently lead users to want to run large numbers of relatively small jobs (often called many-task computing, an ensemble, or a workflow), which can conflict with supercomputer configurations. In this paper, we discuss techniques developed to execute ordinary applications over leadership class supercomputers. We use the high-performance Swift parallel scripting framework and build two workflow execution techniques-sub-jobs and main-wrap. The sub-jobs technique, built on top of the IBM Blue Gene/Q resource manager Cobalt's sub-block jobs, lets users submit multiple, independent, repeated smaller jobs within a single larger resource block. The main-wrap technique is a scheme that enables C/C++ programs to be defined as functions that are wrapped by a high-performance Swift wrapper and that are invoked as a Swift script. We discuss the needs, benefits, technicalities, and current limitations of these techniques. We further discuss the real-world science enabled by these techniques and the results obtained.

Flux-Level Transit Injection Experiments with NASA Pleiades Supercomputer

Science.gov (United States)

Li, Jie; Burke, Christopher J.; Catanzarite, Joseph; Seader, Shawn; Haas, Michael R.; Batalha, Natalie; Henze, Christopher; Christiansen, Jessie; Kepler Project, NASA Advanced Supercomputing Division

2016-06-01

Flux-Level Transit Injection (FLTI) experiments are executed with NASA's Pleiades supercomputer for the Kepler Mission. The latest release (9.3, January 2016) of the Kepler Science Operations Center Pipeline is used in the FLTI experiments. Their purpose is to validate the Analytic Completeness Model (ACM), which can be computed for all Kepler target stars, thereby enabling exoplanet occurrence rate studies. Pleiades, a facility of NASA's Advanced Supercomputing Division, is one of the world's most powerful supercomputers and represents NASA's state-of-the-art technology. We discuss the details of implementing the FLTI experiments on the Pleiades supercomputer. For example, taking into account that ~16 injections are generated by one core of the Pleiades processors in an hour, the “shallow” FLTI experiment, in which ~2000 injections are required per target star, can be done for 16% of all Kepler target stars in about 200 hours. Stripping down the transit search to bare bones, i.e. only searching adjacent high/low periods at high/low pulse durations, makes the computationally intensive FLTI experiments affordable. The design of the FLTI experiments and the analysis of the resulting data are presented in “Validating an Analytic Completeness Model for Kepler Target Stars Based on Flux-level Transit Injection Experiments” by Catanzarite et al. (#2494058).Kepler was selected as the 10th mission of the Discovery Program. Funding for the Kepler Mission has been provided by the NASA Science Mission Directorate.

Extracting the Textual and Temporal Structure of Supercomputing Logs

Energy Technology Data Exchange (ETDEWEB)

Jain, S; Singh, I; Chandra, A; Zhang, Z; Bronevetsky, G

2009-05-26

Supercomputers are prone to frequent faults that adversely affect their performance, reliability and functionality. System logs collected on these systems are a valuable resource of information about their operational status and health. However, their massive size, complexity, and lack of standard format makes it difficult to automatically extract information that can be used to improve system management. In this work we propose a novel method to succinctly represent the contents of supercomputing logs, by using textual clustering to automatically find the syntactic structures of log messages. This information is used to automatically classify messages into semantic groups via an online clustering algorithm. Further, we describe a methodology for using the temporal proximity between groups of log messages to identify correlated events in the system. We apply our proposed methods to two large, publicly available supercomputing logs and show that our technique features nearly perfect accuracy for online log-classification and extracts meaningful structural and temporal message patterns that can be used to improve the accuracy of other log analysis techniques.

Introduction to Reconfigurable Supercomputing

CERN Document Server

Lanzagorta, Marco; Rosenberg, Robert

2010-01-01

This book covers technologies, applications, tools, languages, procedures, advantages, and disadvantages of reconfigurable supercomputing using Field Programmable Gate Arrays (FPGAs). The target audience is the community of users of High Performance Computers (HPe who may benefit from porting their applications into a reconfigurable environment. As such, this book is intended to guide the HPC user through the many algorithmic considerations, hardware alternatives, usability issues, programming languages, and design tools that need to be understood before embarking on the creation of reconfigur

SUPERCOMPUTERS FOR AIDING ECONOMIC PROCESSES WITH REFERENCE TO THE FINANCIAL SECTOR

Directory of Open Access Journals (Sweden)

Jerzy Balicki

2014-12-01

Full Text Available The article discusses the use of supercomputers to support business processes with particular emphasis on the financial sector. A reference was made to the selected projects that support economic development. In particular, we propose the use of supercomputers to perform artificial intel-ligence methods in banking. The proposed methods combined with modern technology enables a significant increase in the competitiveness of enterprises and banks by adding new functionality.

Exploiting Thread Parallelism for Ocean Modeling on Cray XC Supercomputers

Energy Technology Data Exchange (ETDEWEB)

Sarje, Abhinav [Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Jacobsen, Douglas W. [Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Williams, Samuel W. [Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Ringler, Todd [Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Oliker, Leonid [Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)

2016-05-01

The incorporation of increasing core counts in modern processors used to build state-of-the-art supercomputers is driving application development towards exploitation of thread parallelism, in addition to distributed memory parallelism, with the goal of delivering efficient high-performance codes. In this work we describe the exploitation of threading and our experiences with it with respect to a real-world ocean modeling application code, MPAS-Ocean. We present detailed performance analysis and comparisons of various approaches and configurations for threading on the Cray XC series supercomputers.

Visualization environment of the large-scale data of JAEA's supercomputer system

Energy Technology Data Exchange (ETDEWEB)

Sakamoto, Kensaku [Japan Atomic Energy Agency, Center for Computational Science and e-Systems, Tokai, Ibaraki (Japan); Hoshi, Yoshiyuki [Research Organization for Information Science and Technology (RIST), Tokai, Ibaraki (Japan)

2013-11-15

On research and development of various fields of nuclear energy, visualization of calculated data is especially useful to understand the result of simulation in an intuitive way. Many researchers who run simulations on the supercomputer in Japan Atomic Energy Agency (JAEA) are used to transfer calculated data files from the supercomputer to their local PCs for visualization. In recent years, as the size of calculated data has gotten larger with improvement of supercomputer performance, reduction of visualization processing time as well as efficient use of JAEA network is being required. As a solution, we introduced a remote visualization system which has abilities to utilize parallel processors on the supercomputer and to reduce the usage of network resources by transferring data of intermediate visualization process. This paper reports a study on the performance of image processing with the remote visualization system. The visualization processing time is measured and the influence of network speed is evaluated by varying the drawing mode, the size of visualization data and the number of processors. Based on this study, a guideline for using the remote visualization system is provided to show how the system can be used effectively. An upgrade policy of the next system is also shown. (author)

Multi-petascale highly efficient parallel supercomputer

Science.gov (United States)

Asaad, Sameh; Bellofatto, Ralph E.; Blocksome, Michael A.; Blumrich, Matthias A.; Boyle, Peter; Brunheroto, Jose R.; Chen, Dong; Cher, Chen -Yong; Chiu, George L.; Christ, Norman; Coteus, Paul W.; Davis, Kristan D.; Dozsa, Gabor J.; Eichenberger, Alexandre E.; Eisley, Noel A.; Ellavsky, Matthew R.; Evans, Kahn C.; Fleischer, Bruce M.; Fox, Thomas W.; Gara, Alan; Giampapa, Mark E.; Gooding, Thomas M.; Gschwind, Michael K.; Gunnels, John A.; Hall, Shawn A.; Haring, Rudolf A.; Heidelberger, Philip; Inglett, Todd A.; Knudson, Brant L.; Kopcsay, Gerard V.; Kumar, Sameer; Mamidala, Amith R.; Marcella, James A.; Megerian, Mark G.; Miller, Douglas R.; Miller, Samuel J.; Muff, Adam J.; Mundy, Michael B.; O'Brien, John K.; O'Brien, Kathryn M.; Ohmacht, Martin; Parker, Jeffrey J.; Poole, Ruth J.; Ratterman, Joseph D.; Salapura, Valentina; Satterfield, David L.; Senger, Robert M.; Smith, Brian; Steinmacher-Burow, Burkhard; Stockdell, William M.; Stunkel, Craig B.; Sugavanam, Krishnan; Sugawara, Yutaka; Takken, Todd E.; Trager, Barry M.; Van Oosten, James L.; Wait, Charles D.; Walkup, Robert E.; Watson, Alfred T.; Wisniewski, Robert W.; Wu, Peng

2015-07-14

A Multi-Petascale Highly Efficient Parallel Supercomputer of 100 petaOPS-scale computing, at decreased cost, power and footprint, and that allows for a maximum packaging density of processing nodes from an interconnect point of view. The Supercomputer exploits technological advances in VLSI that enables a computing model where many processors can be integrated into a single Application Specific Integrated Circuit (ASIC). Each ASIC computing node comprises a system-on-chip ASIC utilizing four or more processors integrated into one die, with each having full access to all system resources and enabling adaptive partitioning of the processors to functions such as compute or messaging I/O on an application by application basis, and preferably, enable adaptive partitioning of functions in accordance with various algorithmic phases within an application, or if I/O or other processors are underutilized, then can participate in computation or communication nodes are interconnected by a five dimensional torus network with DMA that optimally maximize the throughput of packet communications between nodes and minimize latency.

QCD on the BlueGene/L Supercomputer

International Nuclear Information System (INIS)

Bhanot, G.; Chen, D.; Gara, A.; Sexton, J.; Vranas, P.

2005-01-01

In June 2004 QCD was simulated for the first time at sustained speed exceeding 1 TeraFlops in the BlueGene/L supercomputer at the IBM T.J. Watson Research Lab. The implementation and performance of QCD in the BlueGene/L is presented

QCD on the BlueGene/L Supercomputer

Science.gov (United States)

Bhanot, G.; Chen, D.; Gara, A.; Sexton, J.; Vranas, P.

2005-03-01

In June 2004 QCD was simulated for the first time at sustained speed exceeding 1 TeraFlops in the BlueGene/L supercomputer at the IBM T.J. Watson Research Lab. The implementation and performance of QCD in the BlueGene/L is presented.

Development of seismic tomography software for hybrid supercomputers

Science.gov (United States)

Nikitin, Alexandr; Serdyukov, Alexandr; Duchkov, Anton

2015-04-01

Seismic tomography is a technique used for computing velocity model of geologic structure from first arrival travel times of seismic waves. The technique is used in processing of regional and global seismic data, in seismic exploration for prospecting and exploration of mineral and hydrocarbon deposits, and in seismic engineering for monitoring the condition of engineering structures and the surrounding host medium. As a consequence of development of seismic monitoring systems and increasing volume of seismic data, there is a growing need for new, more effective computational algorithms for use in seismic tomography applications with improved performance, accuracy and resolution. To achieve this goal, it is necessary to use modern high performance computing systems, such as supercomputers with hybrid architecture that use not only CPUs, but also accelerators and co-processors for computation. The goal of this research is the development of parallel seismic tomography algorithms and software package for such systems, to be used in processing of large volumes of seismic data (hundreds of gigabytes and more). These algorithms and software package will be optimized for the most common computing devices used in modern hybrid supercomputers, such as Intel Xeon CPUs, NVIDIA Tesla accelerators and Intel Xeon Phi co-processors. In this work, the following general scheme of seismic tomography is utilized. Using the eikonal equation solver, arrival times of seismic waves are computed based on assumed velocity model of geologic structure being analyzed. In order to solve the linearized inverse problem, tomographic matrix is computed that connects model adjustments with travel time residuals, and the resulting system of linear equations is regularized and solved to adjust the model. The effectiveness of parallel implementations of existing algorithms on target architectures is considered. During the first stage of this work, algorithms were developed for execution on

Proceedings of the first energy research power supercomputer users symposium

International Nuclear Information System (INIS)

1991-01-01

The Energy Research Power Supercomputer Users Symposium was arranged to showcase the richness of science that has been pursued and accomplished in this program through the use of supercomputers and now high performance parallel computers over the last year: this report is the collection of the presentations given at the Symposium. ''Power users'' were invited by the ER Supercomputer Access Committee to show that the use of these computational tools and the associated data communications network, ESNet, go beyond merely speeding up computations. Today the work often directly contributes to the advancement of the conceptual developments in their fields and the computational and network resources form the very infrastructure of today's science. The Symposium also provided an opportunity, which is rare in this day of network access to computing resources, for the invited users to compare and discuss their techniques and approaches with those used in other ER disciplines. The significance of new parallel architectures was highlighted by the interesting evening talk given by Dr. Stephen Orszag of Princeton University

Graphics supercomputer for computational fluid dynamics research

Science.gov (United States)

Liaw, Goang S.

1994-11-01

The objective of this project is to purchase a state-of-the-art graphics supercomputer to improve the Computational Fluid Dynamics (CFD) research capability at Alabama A & M University (AAMU) and to support the Air Force research projects. A cutting-edge graphics supercomputer system, Onyx VTX, from Silicon Graphics Computer Systems (SGI), was purchased and installed. Other equipment including a desktop personal computer, PC-486 DX2 with a built-in 10-BaseT Ethernet card, a 10-BaseT hub, an Apple Laser Printer Select 360, and a notebook computer from Zenith were also purchased. A reading room has been converted to a research computer lab by adding some furniture and an air conditioning unit in order to provide an appropriate working environments for researchers and the purchase equipment. All the purchased equipment were successfully installed and are fully functional. Several research projects, including two existing Air Force projects, are being performed using these facilities.

A workbench for tera-flop supercomputing

International Nuclear Information System (INIS)

Resch, M.M.; Kuester, U.; Mueller, M.S.; Lang, U.

2003-01-01

Supercomputers currently reach a peak performance in the range of TFlop/s. With but one exception - the Japanese Earth Simulator - none of these systems has so far been able to also show a level of sustained performance for a variety of applications that comes close to the peak performance. Sustained TFlop/s are therefore rarely seen. The reasons are manifold and are well known: Bandwidth and latency both for main memory and for the internal network are the key internal technical problems. Cache hierarchies with large caches can bring relief but are no remedy to the problem. However, there are not only technical problems that inhibit the full exploitation by scientists of the potential of modern supercomputers. More and more organizational issues come to the forefront. This paper shows the approach of the High Performance Computing Center Stuttgart (HLRS) to deliver a sustained performance of TFlop/s for a wide range of applications from a large group of users spread over Germany. The core of the concept is the role of the data. Around this we design a simulation workbench that hides the complexity of interacting computers, networks and file systems from the user. (authors)

A visual analytics system for optimizing the performance of large-scale networks in supercomputing systems

Directory of Open Access Journals (Sweden)

Takanori Fujiwara

2018-03-01

Full Text Available The overall efficiency of an extreme-scale supercomputer largely relies on the performance of its network interconnects. Several of the state of the art supercomputers use networks based on the increasingly popular Dragonfly topology. It is crucial to study the behavior and performance of different parallel applications running on Dragonfly networks in order to make optimal system configurations and design choices, such as job scheduling and routing strategies. However, in order to study these temporal network behavior, we would need a tool to analyze and correlate numerous sets of multivariate time-series data collected from the Dragonfly’s multi-level hierarchies. This paper presents such a tool–a visual analytics system–that uses the Dragonfly network to investigate the temporal behavior and optimize the communication performance of a supercomputer. We coupled interactive visualization with time-series analysis methods to help reveal hidden patterns in the network behavior with respect to different parallel applications and system configurations. Our system also provides multiple coordinated views for connecting behaviors observed at different levels of the network hierarchies, which effectively helps visual analysis tasks. We demonstrate the effectiveness of the system with a set of case studies. Our system and findings can not only help improve the communication performance of supercomputing applications, but also the network performance of next-generation supercomputers. Keywords: Supercomputing, Parallel communication network, Dragonfly networks, Time-series data, Performance analysis, Visual analytics

KfK-seminar series on supercomputing und visualization from May till September 1992

International Nuclear Information System (INIS)

Hohenhinnebusch, W.

1993-05-01

During the period of may 1992 to september 1992 a series of seminars was held at KfK on several topics of supercomputing in different fields of application. The aim was to demonstrate the importance of supercomputing and visualization in numerical simulations of complex physical and technical phenomena. This report contains the collection of all submitted seminar papers. (orig./HP) [de

Computational plasma physics and supercomputers. Revision 1

International Nuclear Information System (INIS)

Killeen, J.; McNamara, B.

1985-01-01

The Supercomputers of the 80's are introduced. They are 10 to 100 times more powerful than today's machines. The range of physics modeling in the fusion program is outlined. New machine architecture will influence particular models, but parallel processing poses new programming difficulties. Increasing realism in simulations will require better numerics and more elaborate mathematical models

Application of Supercomputer Technologies for Simulation Of Socio-Economic Systems

Directory of Open Access Journals (Sweden)

Vladimir Valentinovich Okrepilov

2015-06-01

Full Text Available To date, an extensive experience has been accumulated in investigation of problems related to quality, assessment of management systems, modeling of economic system sustainability. The performed studies have created a basis for development of a new research area — Economics of Quality. Its tools allow to use opportunities of model simulation for construction of the mathematical models adequately reflecting the role of quality in natural, technical, social regularities of functioning of the complex socio-economic systems. Extensive application and development of models, and also system modeling with use of supercomputer technologies, on our deep belief, will bring the conducted research of socio-economic systems to essentially new level. Moreover, the current scientific research makes a significant contribution to model simulation of multi-agent social systems and that is not less important, it belongs to the priority areas in development of science and technology in our country. This article is devoted to the questions of supercomputer technologies application in public sciences, first of all, — regarding technical realization of the large-scale agent-focused models (AFM. The essence of this tool is that owing to the power computer increase it has become possible to describe the behavior of many separate fragments of a difficult system, as socio-economic systems are. The article also deals with the experience of foreign scientists and practicians in launching the AFM on supercomputers, and also the example of AFM developed in CEMI RAS, stages and methods of effective calculating kernel display of multi-agent system on architecture of a modern supercomputer will be analyzed. The experiments on the basis of model simulation on forecasting the population of St. Petersburg according to three scenarios as one of the major factors influencing the development of socio-economic system and quality of life of the population are presented in the

Centralized supercomputer support for magnetic fusion energy research

International Nuclear Information System (INIS)

Fuss, D.; Tull, G.G.

1984-01-01

High-speed computers with large memories are vital to magnetic fusion energy research. Magnetohydrodynamic (MHD), transport, equilibrium, Vlasov, particle, and Fokker-Planck codes that model plasma behavior play an important role in designing experimental hardware and interpreting the resulting data, as well as in advancing plasma theory itself. The size, architecture, and software of supercomputers to run these codes are often the crucial constraints on the benefits such computational modeling can provide. Hence, vector computers such as the CRAY-1 offer a valuable research resource. To meet the computational needs of the fusion program, the National Magnetic Fusion Energy Computer Center (NMFECC) was established in 1974 at the Lawrence Livermore National Laboratory. Supercomputers at the central computing facility are linked to smaller computer centers at each of the major fusion laboratories by a satellite communication network. In addition to providing large-scale computing, the NMFECC environment stimulates collaboration and the sharing of computer codes and data among the many fusion researchers in a cost-effective manner

Extending ATLAS Computing to Commercial Clouds and Supercomputers

CERN Document Server

Nilsson, P; The ATLAS collaboration; Filipcic, A; Klimentov, A; Maeno, T; Oleynik, D; Panitkin, S; Wenaus, T; Wu, W

2014-01-01

The Large Hadron Collider will resume data collection in 2015 with substantially increased computing requirements relative to its first 2009-2013 run. A near doubling of the energy and the data rate, high level of event pile-up, and detector upgrades will mean the number and complexity of events to be analyzed will increase dramatically. A naive extrapolation of the Run 1 experience would suggest that a 5-6 fold increase in computing resources are needed - impossible within the anticipated flat computing budgets in the near future. Consequently ATLAS is engaged in an ambitious program to expand its computing to all available resources, notably including opportunistic use of commercial clouds and supercomputers. Such resources present new challenges in managing heterogeneity, supporting data flows, parallelizing workflows, provisioning software, and other aspects of distributed computing, all while minimizing operational load. We will present the ATLAS experience to date with clouds and supercomputers, and des...

Integration Of PanDA Workload Management System With Supercomputers for ATLAS and Data Intensive Science

Energy Technology Data Exchange (ETDEWEB)

De, K [University of Texas at Arlington; Jha, S [Rutgers University; Klimentov, A [Brookhaven National Laboratory (BNL); Maeno, T [Brookhaven National Laboratory (BNL); Nilsson, P [Brookhaven National Laboratory (BNL); Oleynik, D [University of Texas at Arlington; Panitkin, S [Brookhaven National Laboratory (BNL); Wells, Jack C [ORNL; Wenaus, T [Brookhaven National Laboratory (BNL)

2016-01-01

The Large Hadron Collider (LHC), operating at the international CERN Laboratory in Geneva, Switzerland, is leading Big Data driven scientific explorations. Experiments at the LHC explore the fundamental nature of matter and the basic forces that shape our universe, and were recently credited for the discovery of a Higgs boson. ATLAS, one of the largest collaborations ever assembled in the sciences, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. The ATLAS experiment uses PanDA (Production and Data Analysis) Workload Management System for managing the workflow for all data processing on over 150 data centers. Through PanDA, ATLAS physicists see a single computing facility that enables rapid scientific breakthroughs for the experiment, even though the data centers are physically scattered all over the world. While PanDA currently uses more than 250,000 cores with a peak performance of 0.3 petaFLOPS, LHC data taking runs require more resources than Grid computing can possibly provide. To alleviate these challenges, LHC experiments are engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with supercomputers in United States, Europe and Russia (in particular with Titan supercomputer at Oak Ridge Leadership Computing Facility (OLCF), MIRA supercomputer at Argonne Leadership Computing Facilities (ALCF), Supercomputer at the National Research Center Kurchatov Institute , IT4 in Ostrava and others). Current approach utilizes modified PanDA pilot framework for job submission to the supercomputers batch queues and local data management, with light-weight MPI wrappers to run single threaded workloads in parallel on LCFs multi-core worker nodes. This implementation

Tryton Supercomputer Capabilities for Analysis of Massive Data Streams

Directory of Open Access Journals (Sweden)

Krawczyk Henryk

2015-09-01

Full Text Available The recently deployed supercomputer Tryton, located in the Academic Computer Center of Gdansk University of Technology, provides great means for massive parallel processing. Moreover, the status of the Center as one of the main network nodes in the PIONIER network enables the fast and reliable transfer of data produced by miscellaneous devices scattered in the area of the whole country. The typical examples of such data are streams containing radio-telescope and satellite observations. Their analysis, especially with real-time constraints, can be challenging and requires the usage of dedicated software components. We propose a solution for such parallel analysis using the supercomputer, supervised by the KASKADA platform, which with the conjunction with immerse 3D visualization techniques can be used to solve problems such as pulsar detection and chronometric or oil-spill simulation on the sea surface.

Performance modeling of hybrid MPI/OpenMP scientific applications on large-scale multicore supercomputers

KAUST Repository

Wu, Xingfu; Taylor, Valerie

2013-01-01

In this paper, we present a performance modeling framework based on memory bandwidth contention time and a parameterized communication model to predict the performance of OpenMP, MPI and hybrid applications with weak scaling on three large-scale multicore supercomputers: IBM POWER4, POWER5+ and BlueGene/P, and analyze the performance of these MPI, OpenMP and hybrid applications. We use STREAM memory benchmarks and Intel's MPI benchmarks to provide initial performance analysis and model validation of MPI and OpenMP applications on these multicore supercomputers because the measured sustained memory bandwidth can provide insight into the memory bandwidth that a system should sustain on scientific applications with the same amount of workload per core. In addition to using these benchmarks, we also use a weak-scaling hybrid MPI/OpenMP large-scale scientific application: Gyrokinetic Toroidal Code (GTC) in magnetic fusion to validate our performance model of the hybrid application on these multicore supercomputers. The validation results for our performance modeling method show less than 7.77% error rate in predicting the performance of hybrid MPI/OpenMP GTC on up to 512 cores on these multicore supercomputers. © 2013 Elsevier Inc.

Performance modeling of hybrid MPI/OpenMP scientific applications on large-scale multicore supercomputers

KAUST Repository

Wu, Xingfu

2013-12-01

In this paper, we present a performance modeling framework based on memory bandwidth contention time and a parameterized communication model to predict the performance of OpenMP, MPI and hybrid applications with weak scaling on three large-scale multicore supercomputers: IBM POWER4, POWER5+ and BlueGene/P, and analyze the performance of these MPI, OpenMP and hybrid applications. We use STREAM memory benchmarks and Intel\\'s MPI benchmarks to provide initial performance analysis and model validation of MPI and OpenMP applications on these multicore supercomputers because the measured sustained memory bandwidth can provide insight into the memory bandwidth that a system should sustain on scientific applications with the same amount of workload per core. In addition to using these benchmarks, we also use a weak-scaling hybrid MPI/OpenMP large-scale scientific application: Gyrokinetic Toroidal Code (GTC) in magnetic fusion to validate our performance model of the hybrid application on these multicore supercomputers. The validation results for our performance modeling method show less than 7.77% error rate in predicting the performance of hybrid MPI/OpenMP GTC on up to 512 cores on these multicore supercomputers. © 2013 Elsevier Inc.

Guide to dataflow supercomputing basic concepts, case studies, and a detailed example

CERN Document Server

Milutinovic, Veljko; Trifunovic, Nemanja; Giorgi, Roberto

2015-01-01

This unique text/reference describes an exciting and novel approach to supercomputing in the DataFlow paradigm. The major advantages and applications of this approach are clearly described, and a detailed explanation of the programming model is provided using simple yet effective examples. The work is developed from a series of lecture courses taught by the authors in more than 40 universities across more than 20 countries, and from research carried out by Maxeler Technologies, Inc. Topics and features: presents a thorough introduction to DataFlow supercomputing for big data problems; revie

Enabling Diverse Software Stacks on Supercomputers using High Performance Virtual Clusters.

Energy Technology Data Exchange (ETDEWEB)

Younge, Andrew J. [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); Pedretti, Kevin [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); Grant, Ryan [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); Brightwell, Ron [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)

2017-05-01

While large-scale simulations have been the hallmark of the High Performance Computing (HPC) community for decades, Large Scale Data Analytics (LSDA) workloads are gaining attention within the scientific community not only as a processing component to large HPC simulations, but also as standalone scientific tools for knowledge discovery. With the path towards Exascale, new HPC runtime systems are also emerging in a way that differs from classical distributed com- puting models. However, system software for such capabilities on the latest extreme-scale DOE supercomputing needs to be enhanced to more appropriately support these types of emerging soft- ware ecosystems. In this paper, we propose the use of Virtual Clusters on advanced supercomputing resources to enable systems to support not only HPC workloads, but also emerging big data stacks. Specifi- cally, we have deployed the KVM hypervisor within Cray's Compute Node Linux on a XC-series supercomputer testbed. We also use libvirt and QEMU to manage and provision VMs directly on compute nodes, leveraging Ethernet-over-Aries network emulation. To our knowledge, this is the first known use of KVM on a true MPP supercomputer. We investigate the overhead our solution using HPC benchmarks, both evaluating single-node performance as well as weak scaling of a 32-node virtual cluster. Overall, we find single node performance of our solution using KVM on a Cray is very efficient with near-native performance. However overhead increases by up to 20% as virtual cluster size increases, due to limitations of the Ethernet-over-Aries bridged network. Furthermore, we deploy Apache Spark with large data analysis workloads in a Virtual Cluster, ef- fectively demonstrating how diverse software ecosystems can be supported by High Performance Virtual Clusters.

Cellular-automata supercomputers for fluid-dynamics modeling

International Nuclear Information System (INIS)

Margolus, N.; Toffoli, T.; Vichniac, G.

1986-01-01

We report recent developments in the modeling of fluid dynamics, and give experimental results (including dynamical exponents) obtained using cellular automata machines. Because of their locality and uniformity, cellular automata lend themselves to an extremely efficient physical realization; with a suitable architecture, an amount of hardware resources comparable to that of a home computer can achieve (in the simulation of cellular automata) the performance of a conventional supercomputer

The TeraGyroid Experiment – Supercomputing 2003

Directory of Open Access Journals (Sweden)

R.J. Blake

2005-01-01

Full Text Available Amphiphiles are molecules with hydrophobic tails and hydrophilic heads. When dispersed in solvents, they self assemble into complex mesophases including the beautiful cubic gyroid phase. The goal of the TeraGyroid experiment was to study defect pathways and dynamics in these gyroids. The UK's supercomputing and USA's TeraGrid facilities were coupled together, through a dedicated high-speed network, into a single computational Grid for research work that peaked around the Supercomputing 2003 conference. The gyroids were modeled using lattice Boltzmann methods with parameter spaces explored using many 1283 and 3grid point simulations, this data being used to inform the world's largest three-dimensional time dependent simulation with 10243-grid points. The experiment generated some 2 TBytes of useful data. In terms of Grid technology the project demonstrated the migration of simulations (using Globus middleware to and fro across the Atlantic exploiting the availability of resources. Integration of the systems accelerated the time to insight. Distributed visualisation of the output datasets enabled the parameter space of the interactions within the complex fluid to be explored from a number of sites, informed by discourse over the Access Grid. The project was sponsored by EPSRC (UK and NSF (USA with trans-Atlantic optical bandwidth provided by British Telecommunications.

Integration Of PanDA Workload Management System With Supercomputers for ATLAS and Data Intensive Science

Science.gov (United States)

Klimentov, A.; De, K.; Jha, S.; Maeno, T.; Nilsson, P.; Oleynik, D.; Panitkin, S.; Wells, J.; Wenaus, T.

2016-10-01

The.LHC, operating at CERN, is leading Big Data driven scientific explorations. Experiments at the LHC explore the fundamental nature of matter and the basic forces that shape our universe. ATLAS, one of the largest collaborations ever assembled in the sciences, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. The ATLAS experiment uses PanDA (Production and Data Analysis) Workload Management System for managing the workflow for all data processing on over 150 data centers. Through PanDA, ATLAS physicists see a single computing facility that enables rapid scientific breakthroughs for the experiment, even though the data centers are physically scattered all over the world. While PanDA currently uses more than 250,000 cores with a peak performance of 0.3 petaFLOPS, LHC data taking runs require more resources than grid can possibly provide. To alleviate these challenges, LHC experiments are engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with supercomputers in United States, in particular with Titan supercomputer at Oak Ridge Leadership Computing Facility. Current approach utilizes modified PanDA pilot framework for job submission to the supercomputers batch queues and local data management, with light-weight MPI wrappers to run single threaded workloads in parallel on LCFs multi-core worker nodes. This implementation was tested with a variety of Monte-Carlo workloads on several supercomputing platforms for ALICE and ATLAS experiments and it is in full pro duction for the ATLAS since September 2015. We will present our current accomplishments with running PanDA at supercomputers and demonstrate our ability to use PanDA as a portal independent of the

Integration Of PanDA Workload Management System With Supercomputers for ATLAS and Data Intensive Science

International Nuclear Information System (INIS)

Klimentov, A; Maeno, T; Nilsson, P; Panitkin, S; Wenaus, T; De, K; Oleynik, D; Jha, S; Wells, J

2016-01-01

The.LHC, operating at CERN, is leading Big Data driven scientific explorations. Experiments at the LHC explore the fundamental nature of matter and the basic forces that shape our universe. ATLAS, one of the largest collaborations ever assembled in the sciences, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. The ATLAS experiment uses PanDA (Production and Data Analysis) Workload Management System for managing the workflow for all data processing on over 150 data centers. Through PanDA, ATLAS physicists see a single computing facility that enables rapid scientific breakthroughs for the experiment, even though the data centers are physically scattered all over the world. While PanDA currently uses more than 250,000 cores with a peak performance of 0.3 petaFLOPS, LHC data taking runs require more resources than grid can possibly provide. To alleviate these challenges, LHC experiments are engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with supercomputers in United States, in particular with Titan supercomputer at Oak Ridge Leadership Computing Facility. Current approach utilizes modified PanDA pilot framework for job submission to the supercomputers batch queues and local data management, with light-weight MPI wrappers to run single threaded workloads in parallel on LCFs multi-core worker nodes. This implementation was tested with a variety of Monte-Carlo workloads on several supercomputing platforms for ALICE and ATLAS experiments and it is in full pro duction for the ATLAS since September 2015. We will present our current accomplishments with running PanDA at supercomputers and demonstrate our ability to use PanDA as a portal independent of the

Analyzing the Interplay of Failures and Workload on a Leadership-Class Supercomputer

Energy Technology Data Exchange (ETDEWEB)

Meneses, Esteban [University of Pittsburgh; Ni, Xiang [University of Illinois at Urbana-Champaign; Jones, Terry R [ORNL; Maxwell, Don E [ORNL

2015-01-01

The unprecedented computational power of cur- rent supercomputers now makes possible the exploration of complex problems in many scientific fields, from genomic analysis to computational fluid dynamics. Modern machines are powerful because they are massive: they assemble millions of cores and a huge quantity of disks, cards, routers, and other components. But it is precisely the size of these machines that glooms the future of supercomputing. A system that comprises many components has a high chance to fail, and fail often. In order to make the next generation of supercomputers usable, it is imperative to use some type of fault tolerance platform to run applications on large machines. Most fault tolerance strategies can be optimized for the peculiarities of each system and boost efficacy by keeping the system productive. In this paper, we aim to understand how failure characterization can improve resilience in several layers of the software stack: applications, runtime systems, and job schedulers. We examine the Titan supercomputer, one of the fastest systems in the world. We analyze a full year of Titan in production and distill the failure patterns of the machine. By looking into Titan s log files and using the criteria of experts, we provide a detailed description of the types of failures. In addition, we inspect the job submission files and describe how the system is used. Using those two sources, we cross correlate failures in the machine to executing jobs and provide a picture of how failures affect the user experience. We believe such characterization is fundamental in developing appropriate fault tolerance solutions for Cray systems similar to Titan.

Role of supercomputers in magnetic fusion and energy research programs

International Nuclear Information System (INIS)

Killeen, J.

1985-06-01

The importance of computer modeling in magnetic fusion (MFE) and energy research (ER) programs is discussed. The need for the most advanced supercomputers is described, and the role of the National Magnetic Fusion Energy Computer Center in meeting these needs is explained

Supercomputer and cluster performance modeling and analysis efforts:2004-2006.

Energy Technology Data Exchange (ETDEWEB)

Sturtevant, Judith E.; Ganti, Anand; Meyer, Harold (Hal) Edward; Stevenson, Joel O.; Benner, Robert E., Jr. (.,; .); Goudy, Susan Phelps; Doerfler, Douglas W.; Domino, Stefan Paul; Taylor, Mark A.; Malins, Robert Joseph; Scott, Ryan T.; Barnette, Daniel Wayne; Rajan, Mahesh; Ang, James Alfred; Black, Amalia Rebecca; Laub, Thomas William; Vaughan, Courtenay Thomas; Franke, Brian Claude

2007-02-01

This report describes efforts by the Performance Modeling and Analysis Team to investigate performance characteristics of Sandia's engineering and scientific applications on the ASC capability and advanced architecture supercomputers, and Sandia's capacity Linux clusters. Efforts to model various aspects of these computers are also discussed. The goals of these efforts are to quantify and compare Sandia's supercomputer and cluster performance characteristics; to reveal strengths and weaknesses in such systems; and to predict performance characteristics of, and provide guidelines for, future acquisitions and follow-on systems. Described herein are the results obtained from running benchmarks and applications to extract performance characteristics and comparisons, as well as modeling efforts, obtained during the time period 2004-2006. The format of the report, with hypertext links to numerous additional documents, purposefully minimizes the document size needed to disseminate the extensive results from our research.

Ultrascalable petaflop parallel supercomputer

Science.gov (United States)

Blumrich, Matthias A [Ridgefield, CT; Chen, Dong [Croton On Hudson, NY; Chiu, George [Cross River, NY; Cipolla, Thomas M [Katonah, NY; Coteus, Paul W [Yorktown Heights, NY; Gara, Alan G [Mount Kisco, NY; Giampapa, Mark E [Irvington, NY; Hall, Shawn [Pleasantville, NY; Haring, Rudolf A [Cortlandt Manor, NY; Heidelberger, Philip [Cortlandt Manor, NY; Kopcsay, Gerard V [Yorktown Heights, NY; Ohmacht, Martin [Yorktown Heights, NY; Salapura, Valentina [Chappaqua, NY; Sugavanam, Krishnan [Mahopac, NY; Takken, Todd [Brewster, NY

2010-07-20

A massively parallel supercomputer of petaOPS-scale includes node architectures based upon System-On-a-Chip technology, where each processing node comprises a single Application Specific Integrated Circuit (ASIC) having up to four processing elements. The ASIC nodes are interconnected by multiple independent networks that optimally maximize the throughput of packet communications between nodes with minimal latency. The multiple networks may include three high-speed networks for parallel algorithm message passing including a Torus, collective network, and a Global Asynchronous network that provides global barrier and notification functions. These multiple independent networks may be collaboratively or independently utilized according to the needs or phases of an algorithm for optimizing algorithm processing performance. The use of a DMA engine is provided to facilitate message passing among the nodes without the expenditure of processing resources at the node.

Direct exploitation of a top 500 Supercomputer for Analysis of CMS Data

International Nuclear Information System (INIS)

Cabrillo, I; Cabellos, L; Marco, J; Fernandez, J; Gonzalez, I

2014-01-01

The Altamira Supercomputer hosted at the Instituto de Fisica de Cantatbria (IFCA) entered in operation in summer 2012. Its last generation FDR Infiniband network used (for message passing) in parallel jobs, supports the connection to General Parallel File System (GPFS) servers, enabling an efficient simultaneous processing of multiple data demanding jobs. Sharing a common GPFS system and a single LDAP-based identification with the existing Grid clusters at IFCA allows CMS researchers to exploit the large instantaneous capacity of this supercomputer to execute analysis jobs. The detailed experience describing this opportunistic use for skimming and final analysis of CMS 2012 data for a specific physics channel, resulting in an order of magnitude reduction of the waiting time, is presented.

Toward a Proof of Concept Cloud Framework for Physics Applications on Blue Gene Supercomputers

International Nuclear Information System (INIS)

Dreher, Patrick; Scullin, William; Vouk, Mladen

2015-01-01

Traditional high performance supercomputers are capable of delivering large sustained state-of-the-art computational resources to physics applications over extended periods of time using batch processing mode operating environments. However, today there is an increasing demand for more complex workflows that involve large fluctuations in the levels of HPC physics computational requirements during the simulations. Some of the workflow components may also require a richer set of operating system features and schedulers than normally found in a batch oriented HPC environment. This paper reports on progress toward a proof of concept design that implements a cloud framework onto BG/P and BG/Q platforms at the Argonne Leadership Computing Facility. The BG/P implementation utilizes the Kittyhawk utility and the BG/Q platform uses an experimental heterogeneous FusedOS operating system environment. Both platforms use the Virtual Computing Laboratory as the cloud computing system embedded within the supercomputer. This proof of concept design allows a cloud to be configured so that it can capitalize on the specialized infrastructure capabilities of a supercomputer and the flexible cloud configurations without resorting to virtualization. Initial testing of the proof of concept system is done using the lattice QCD MILC code. These types of user reconfigurable environments have the potential to deliver experimental schedulers and operating systems within a working HPC environment for physics computations that may be different from the native OS and schedulers on production HPC supercomputers. (paper)

Plane-wave electronic structure calculations on a parallel supercomputer

International Nuclear Information System (INIS)

Nelson, J.S.; Plimpton, S.J.; Sears, M.P.

1993-01-01

The development of iterative solutions of Schrodinger's equation in a plane-wave (pw) basis over the last several years has coincided with great advances in the computational power available for performing the calculations. These dual developments have enabled many new and interesting condensed matter phenomena to be studied from a first-principles approach. The authors present a detailed description of the implementation on a parallel supercomputer (hypercube) of the first-order equation-of-motion solution to Schrodinger's equation, using plane-wave basis functions and ab initio separable pseudopotentials. By distributing the plane-waves across the processors of the hypercube many of the computations can be performed in parallel, resulting in decreases in the overall computation time relative to conventional vector supercomputers. This partitioning also provides ample memory for large Fast Fourier Transform (FFT) meshes and the storage of plane-wave coefficients for many hundreds of energy bands. The usefulness of the parallel techniques is demonstrated by benchmark timings for both the FFT's and iterations of the self-consistent solution of Schrodinger's equation for different sized Si unit cells of up to 512 atoms

Problem solving in nuclear engineering using supercomputers

International Nuclear Information System (INIS)

Schmidt, F.; Scheuermann, W.; Schatz, A.

1987-01-01

The availability of supercomputers enables the engineer to formulate new strategies for problem solving. One such strategy is the Integrated Planning and Simulation System (IPSS). With the integrated systems, simulation models with greater consistency and good agreement with actual plant data can be effectively realized. In the present work some of the basic ideas of IPSS are described as well as some of the conditions necessary to build such systems. Hardware and software characteristics as realized are outlined. (orig.) [de

FPS scientific and supercomputers computers in chemistry

International Nuclear Information System (INIS)

Curington, I.J.

1987-01-01

FPS Array Processors, scientific computers, and highly parallel supercomputers are used in nearly all aspects of compute-intensive computational chemistry. A survey is made of work utilizing this equipment, both published and current research. The relationship of the computer architecture to computational chemistry is discussed, with specific reference to Molecular Dynamics, Quantum Monte Carlo simulations, and Molecular Graphics applications. Recent installations of the FPS T-Series are highlighted, and examples of Molecular Graphics programs running on the FPS-5000 are shown

Visualizing quantum scattering on the CM-2 supercomputer

International Nuclear Information System (INIS)

Richardson, J.L.

1991-01-01

We implement parallel algorithms for solving the time-dependent Schroedinger equation on the CM-2 supercomputer. These methods are unconditionally stable as well as unitary at each time step and have the advantage of being spatially local and explicit. We show how to visualize the dynamics of quantum scattering using techniques for visualizing complex wave functions. Several scattering problems are solved to demonstrate the use of these methods. (orig.)

Integration of Titan supercomputer at OLCF with ATLAS Production System

CERN Document Server

AUTHOR|(SzGeCERN)643806; The ATLAS collaboration; De, Kaushik; Klimentov, Alexei; Nilsson, Paul; Oleynik, Danila; Padolski, Siarhei; Panitkin, Sergey; Wenaus, Torre

2017-01-01

The PanDA (Production and Distributed Analysis) workload management system was developed to meet the scale and complexity of distributed computing for the ATLAS experiment. PanDA managed resources are distributed worldwide, on hundreds of computing sites, with thousands of physicists accessing hundreds of Petabytes of data and the rate of data processing already exceeds Exabyte per year. While PanDA currently uses more than 200,000 cores at well over 100 Grid sites, future LHC data taking runs will require more resources than Grid computing can possibly provide. Additional computing and storage resources are required. Therefore ATLAS is engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. In this paper we will describe a project aimed at integration of ATLAS Production System with Titan supercomputer at Oak Ridge Leadership Computing Facility (OLCF). Current approach utilizes modified PanDA Pilot framework for jo...

Integration of Titan supercomputer at OLCF with ATLAS production system

CERN Document Server

Panitkin, Sergey; The ATLAS collaboration

2016-01-01

The PanDA (Production and Distributed Analysis) workload management system was developed to meet the scale and complexity of distributed computing for the ATLAS experiment. PanDA managed resources are distributed worldwide, on hundreds of computing sites, with thousands of physicists accessing hundreds of Petabytes of data and the rate of data processing already exceeds Exabyte per year. While PanDA currently uses more than 200,000 cores at well over 100 Grid sites, future LHC data taking runs will require more resources than Grid computing can possibly provide. Additional computing and storage resources are required. Therefore ATLAS is engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. In this talk we will describe a project aimed at integration of ATLAS Production System with Titan supercomputer at Oak Ridge Leadership Computing Facility (OLCF). Current approach utilizes modified PanDA Pilot framework for job...

Supercomputer algorithms for reactivity, dynamics and kinetics of small molecules

International Nuclear Information System (INIS)

Lagana, A.

1989-01-01

Even for small systems, the accurate characterization of reactive processes is so demanding of computer resources as to suggest the use of supercomputers having vector and parallel facilities. The full advantages of vector and parallel architectures can sometimes be obtained by simply modifying existing programs, vectorizing the manipulation of vectors and matrices, and requiring the parallel execution of independent tasks. More often, however, a significant time saving can be obtained only when the computer code undergoes a deeper restructuring, requiring a change in the computational strategy or, more radically, the adoption of a different theoretical treatment. This book discusses supercomputer strategies based upon act and approximate methods aimed at calculating the electronic structure and the reactive properties of small systems. The book shows how, in recent years, intense design activity has led to the ability to calculate accurate electronic structures for reactive systems, exact and high-level approximations to three-dimensional reactive dynamics, and to efficient directive and declaratory software for the modelling of complex systems

Design of multiple sequence alignment algorithms on parallel, distributed memory supercomputers.

Science.gov (United States)

Church, Philip C; Goscinski, Andrzej; Holt, Kathryn; Inouye, Michael; Ghoting, Amol; Makarychev, Konstantin; Reumann, Matthias

2011-01-01

The challenge of comparing two or more genomes that have undergone recombination and substantial amounts of segmental loss and gain has recently been addressed for small numbers of genomes. However, datasets of hundreds of genomes are now common and their sizes will only increase in the future. Multiple sequence alignment of hundreds of genomes remains an intractable problem due to quadratic increases in compute time and memory footprint. To date, most alignment algorithms are designed for commodity clusters without parallelism. Hence, we propose the design of a multiple sequence alignment algorithm on massively parallel, distributed memory supercomputers to enable research into comparative genomics on large data sets. Following the methodology of the sequential progressiveMauve algorithm, we design data structures including sequences and sorted k-mer lists on the IBM Blue Gene/P supercomputer (BG/P). Preliminary results show that we can reduce the memory footprint so that we can potentially align over 250 bacterial genomes on a single BG/P compute node. We verify our results on a dataset of E.coli, Shigella and S.pneumoniae genomes. Our implementation returns results matching those of the original algorithm but in 1/2 the time and with 1/4 the memory footprint for scaffold building. In this study, we have laid the basis for multiple sequence alignment of large-scale datasets on a massively parallel, distributed memory supercomputer, thus enabling comparison of hundreds instead of a few genome sequences within reasonable time.

Novel Supercomputing Approaches for High Performance Linear Algebra Using FPGAs, Phase II

Data.gov (United States)

National Aeronautics and Space Administration — Supercomputing plays a major role in many areas of science and engineering, and it has had tremendous impact for decades in areas such as aerospace, defense, energy,...

BSMBench: a flexible and scalable supercomputer benchmark from computational particle physics

CERN Document Server

Bennett, Ed; Del Debbio, Luigi; Jordan, Kirk; Patella, Agostino; Pica, Claudio; Rago, Antonio

2016-01-01

Benchmarking plays a central role in the evaluation of High Performance Computing architectures. Several benchmarks have been designed that allow users to stress various components of supercomputers. In order for the figures they provide to be useful, benchmarks need to be representative of the most common real-world scenarios. In this work, we introduce BSMBench, a benchmarking suite derived from Monte Carlo code used in computational particle physics. The advantage of this suite (which can be freely downloaded from http://www.bsmbench.org/) over others is the capacity to vary the relative importance of computation and communication. This enables the tests to simulate various practical situations. To showcase BSMBench, we perform a wide range of tests on various architectures, from desktop computers to state-of-the-art supercomputers, and discuss the corresponding results. Possible future directions of development of the benchmark are also outlined.

High Performance Networks From Supercomputing to Cloud Computing

CERN Document Server

Abts, Dennis

2011-01-01

Datacenter networks provide the communication substrate for large parallel computer systems that form the ecosystem for high performance computing (HPC) systems and modern Internet applications. The design of new datacenter networks is motivated by an array of applications ranging from communication intensive climatology, complex material simulations and molecular dynamics to such Internet applications as Web search, language translation, collaborative Internet applications, streaming video and voice-over-IP. For both Supercomputing and Cloud Computing the network enables distributed applicati

Intelligent Personal Supercomputer for Solving Scientific and Technical Problems

Directory of Open Access Journals (Sweden)

Khimich, O.M.

2016-09-01

Full Text Available New domestic intellіgent personal supercomputer of hybrid architecture Inparkom_pg for the mathematical modeling of processes in the defense industry, engineering, construction, etc. was developed. Intelligent software for the automatic research and tasks of computational mathematics with approximate data of different structures was designed. Applied software to provide mathematical modeling problems in construction, welding and filtration processes was implemented.

Supercomputers and the future of computational atomic scattering physics

International Nuclear Information System (INIS)

Younger, S.M.

1989-01-01

The advent of the supercomputer has opened new vistas for the computational atomic physicist. Problems of hitherto unparalleled complexity are now being examined using these new machines, and important connections with other fields of physics are being established. This talk briefly reviews some of the most important trends in computational scattering physics and suggests some exciting possibilities for the future. 7 refs., 2 figs

Visualization on supercomputing platform level II ASC milestone (3537-1B) results from Sandia.

Energy Technology Data Exchange (ETDEWEB)

Geveci, Berk (Kitware, Inc., Clifton Park, NY); Fabian, Nathan; Marion, Patrick (Kitware, Inc., Clifton Park, NY); Moreland, Kenneth D.

2010-09-01

This report provides documentation for the completion of the Sandia portion of the ASC Level II Visualization on the platform milestone. This ASC Level II milestone is a joint milestone between Sandia National Laboratories and Los Alamos National Laboratories. This milestone contains functionality required for performing visualization directly on a supercomputing platform, which is necessary for peta-scale visualization. Sandia's contribution concerns in-situ visualization, running a visualization in tandem with a solver. Visualization and analysis of petascale data is limited by several factors which must be addressed as ACES delivers the Cielo platform. Two primary difficulties are: (1) Performance of interactive rendering, which is most computationally intensive portion of the visualization process. For terascale platforms, commodity clusters with graphics processors(GPUs) have been used for interactive rendering. For petascale platforms, visualization and rendering may be able to run efficiently on the supercomputer platform itself. (2) I/O bandwidth, which limits how much information can be written to disk. If we simply analyze the sparse information that is saved to disk we miss the opportunity to analyze the rich information produced every timestep by the simulation. For the first issue, we are pursuing in-situ analysis, in which simulations are coupled directly with analysis libraries at runtime. This milestone will evaluate the visualization and rendering performance of current and next generation supercomputers in contrast to GPU-based visualization clusters, and evaluate the performance of common analysis libraries coupled with the simulation that analyze and write data to disk during a running simulation. This milestone will explore, evaluate and advance the maturity level of these technologies and their applicability to problems of interest to the ASC program. Scientific simulation on parallel supercomputers is traditionally performed in four

Multi-petascale highly efficient parallel supercomputer

Science.gov (United States)

Asaad, Sameh; Bellofatto, Ralph E.; Blocksome, Michael A.; Blumrich, Matthias A.; Boyle, Peter; Brunheroto, Jose R.; Chen, Dong; Cher, Chen-Yong; Chiu, George L.; Christ, Norman; Coteus, Paul W.; Davis, Kristan D.; Dozsa, Gabor J.; Eichenberger, Alexandre E.; Eisley, Noel A.; Ellavsky, Matthew R.; Evans, Kahn C.; Fleischer, Bruce M.; Fox, Thomas W.; Gara, Alan; Giampapa, Mark E.; Gooding, Thomas M.; Gschwind, Michael K.; Gunnels, John A.; Hall, Shawn A.; Haring, Rudolf A.; Heidelberger, Philip; Inglett, Todd A.; Knudson, Brant L.; Kopcsay, Gerard V.; Kumar, Sameer; Mamidala, Amith R.; Marcella, James A.; Megerian, Mark G.; Miller, Douglas R.; Miller, Samuel J.; Muff, Adam J.; Mundy, Michael B.; O'Brien, John K.; O'Brien, Kathryn M.; Ohmacht, Martin; Parker, Jeffrey J.; Poole, Ruth J.; Ratterman, Joseph D.; Salapura, Valentina; Satterfield, David L.; Senger, Robert M.; Steinmacher-Burow, Burkhard; Stockdell, William M.; Stunkel, Craig B.; Sugavanam, Krishnan; Sugawara, Yutaka; Takken, Todd E.; Trager, Barry M.; Van Oosten, James L.; Wait, Charles D.; Walkup, Robert E.; Watson, Alfred T.; Wisniewski, Robert W.; Wu, Peng

2018-05-15

A Multi-Petascale Highly Efficient Parallel Supercomputer of 100 petaflop-scale includes node architectures based upon System-On-a-Chip technology, where each processing node comprises a single Application Specific Integrated Circuit (ASIC). The ASIC nodes are interconnected by a five dimensional torus network that optimally maximize the throughput of packet communications between nodes and minimize latency. The network implements collective network and a global asynchronous network that provides global barrier and notification functions. Integrated in the node design include a list-based prefetcher. The memory system implements transaction memory, thread level speculation, and multiversioning cache that improves soft error rate at the same time and supports DMA functionality allowing for parallel processing message-passing.

Mathematical methods and supercomputing in nuclear applications. Proceedings. Vol. 2

International Nuclear Information System (INIS)

Kuesters, H.; Stein, E.; Werner, W.

1993-04-01

All papers of the two volumes are separately indexed in the data base. Main topics are: Progress in advanced numerical techniques, fluid mechanics, on-line systems, artificial intelligence applications, nodal methods reactor kinetics, reactor design, supercomputer architecture, probabilistic estimation of risk assessment, methods in transport theory, advances in Monte Carlo techniques, and man-machine interface. (orig.)

Mathematical methods and supercomputing in nuclear applications. Proceedings. Vol. 1

International Nuclear Information System (INIS)

Kuesters, H.; Stein, E.; Werner, W.

1993-04-01

All papers of the two volumes are separately indexed in the data base. Main topics are: Progress in advanced numerical techniques, fluid mechanics, on-line systems, artificial intelligence applications, nodal methods reactor kinetics, reactor design, supercomputer architecture, probabilistic estimation of risk assessment, methods in transport theory, advances in Monte Carlo techniques, and man-machine interface. (orig.)

Personal Supercomputing for Monte Carlo Simulation Using a GPU

Energy Technology Data Exchange (ETDEWEB)

Oh, Jae-Yong; Koo, Yang-Hyun; Lee, Byung-Ho [Korea Atomic Energy Research Institute, Daejeon (Korea, Republic of)

2008-05-15

Since the usability, accessibility, and maintenance of a personal computer (PC) are very good, a PC is a useful computer simulation tool for researchers. It has enough calculation power to simulate a small scale system with the improved performance of a PC's CPU. However, if a system is large or long time scale, we need a cluster computer or supercomputer. Recently great changes have occurred in the PC calculation environment. A graphic process unit (GPU) on a graphic card, only used to calculate display data, has a superior calculation capability to a PC's CPU. This GPU calculation performance is a match for the supercomputer in 2000. Although it has such a great calculation potential, it is not easy to program a simulation code for GPU due to difficult programming techniques for converting a calculation matrix to a 3D rendering image using graphic APIs. In 2006, NVIDIA provided the Software Development Kit (SDK) for the programming environment for NVIDIA's graphic cards, which is called the Compute Unified Device Architecture (CUDA). It makes the programming on the GPU easy without knowledge of the graphic APIs. This paper describes the basic architectures of NVIDIA's GPU and CUDA, and carries out a performance benchmark for the Monte Carlo simulation.

Personal Supercomputing for Monte Carlo Simulation Using a GPU

International Nuclear Information System (INIS)

Oh, Jae-Yong; Koo, Yang-Hyun; Lee, Byung-Ho

2008-01-01

Since the usability, accessibility, and maintenance of a personal computer (PC) are very good, a PC is a useful computer simulation tool for researchers. It has enough calculation power to simulate a small scale system with the improved performance of a PC's CPU. However, if a system is large or long time scale, we need a cluster computer or supercomputer. Recently great changes have occurred in the PC calculation environment. A graphic process unit (GPU) on a graphic card, only used to calculate display data, has a superior calculation capability to a PC's CPU. This GPU calculation performance is a match for the supercomputer in 2000. Although it has such a great calculation potential, it is not easy to program a simulation code for GPU due to difficult programming techniques for converting a calculation matrix to a 3D rendering image using graphic APIs. In 2006, NVIDIA provided the Software Development Kit (SDK) for the programming environment for NVIDIA's graphic cards, which is called the Compute Unified Device Architecture (CUDA). It makes the programming on the GPU easy without knowledge of the graphic APIs. This paper describes the basic architectures of NVIDIA's GPU and CUDA, and carries out a performance benchmark for the Monte Carlo simulation

Design and performance characterization of electronic structure calculations on massively parallel supercomputers

DEFF Research Database (Denmark)

Romero, N. A.; Glinsvad, Christian; Larsen, Ask Hjorth

2013-01-01

Density function theory (DFT) is the most widely employed electronic structure method because of its favorable scaling with system size and accuracy for a broad range of molecular and condensed-phase systems. The advent of massively parallel supercomputers has enhanced the scientific community...

Computational Science with the Titan Supercomputer: Early Outcomes and Lessons Learned

Science.gov (United States)

Wells, Jack

2014-03-01

Modeling and simulation with petascale computing has supercharged the process of innovation and understanding, dramatically accelerating time-to-insight and time-to-discovery. This presentation will focus on early outcomes from the Titan supercomputer at the Oak Ridge National Laboratory. Titan has over 18,000 hybrid compute nodes consisting of both CPUs and GPUs. In this presentation, I will discuss the lessons we have learned in deploying Titan and preparing applications to move from conventional CPU architectures to a hybrid machine. I will present early results of materials applications running on Titan and the implications for the research community as we prepare for exascale supercomputer in the next decade. Lastly, I will provide an overview of user programs at the Oak Ridge Leadership Computing Facility with specific information how researchers may apply for allocations of computing resources. This research used resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725.

Integration of PanDA workload management system with Titan supercomputer at OLCF

CERN Document Server

AUTHOR|(INSPIRE)INSPIRE-00300320; Klimentov, Alexei; Oleynik, Danila; Panitkin, Sergey; Petrosyan, Artem; Vaniachine, Alexandre; Wenaus, Torre; Schovancova, Jaroslava

2015-01-01

The PanDA (Production and Distributed Analysis) workload management system (WMS) was developed to meet the scale and complexity of LHC distributed computing for the ATLAS experiment. While PanDA currently distributes jobs to more than 100,000 cores at well over 100 Grid sites, next LHC data taking run will require more resources than Grid computing can possibly provide. To alleviate these challenges, ATLAS is engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with Titan supercomputer at Oak Ridge Leadership Computing Facility (OLCF). Current approach utilizes modi ed PanDA pilot framework for job submission to Titan's batch queues and local data management, with light-weight MPI wrappers to run single threaded workloads in parallel on Titan's multi-core worker nodes. It also gives PanDA new capability to collect, in real time, information about unused...

Integration of PanDA workload management system with Titan supercomputer at OLCF

CERN Document Server

Panitkin, Sergey; The ATLAS collaboration; Klimentov, Alexei; Oleynik, Danila; Petrosyan, Artem; Schovancova, Jaroslava; Vaniachine, Alexandre; Wenaus, Torre

2015-01-01

The PanDA (Production and Distributed Analysis) workload management system (WMS) was developed to meet the scale and complexity of LHC distributed computing for the ATLAS experiment. While PanDA currently uses more than 100,000 cores at well over 100 Grid sites with a peak performance of 0.3 petaFLOPS, next LHC data taking run will require more resources than Grid computing can possibly provide. To alleviate these challenges, ATLAS is engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with Titan supercomputer at Oak Ridge Leadership Computing Facility (OLCF). Current approach utilizes modified PanDA pilot framework for job submission to Titan's batch queues and local data management, with light-weight MPI wrappers to run single threaded workloads in parallel on Titan's multi-core worker nodes. It also gives PanDA new capability to collect, in real tim...

Computational fluid dynamics research at the United Technologies Research Center requiring supercomputers

Science.gov (United States)

Landgrebe, Anton J.

1987-01-01

An overview of research activities at the United Technologies Research Center (UTRC) in the area of Computational Fluid Dynamics (CFD) is presented. The requirement and use of various levels of computers, including supercomputers, for the CFD activities is described. Examples of CFD directed toward applications to helicopters, turbomachinery, heat exchangers, and the National Aerospace Plane are included. Helicopter rotor codes for the prediction of rotor and fuselage flow fields and airloads were developed with emphasis on rotor wake modeling. Airflow and airload predictions and comparisons with experimental data are presented. Examples are presented of recent parabolized Navier-Stokes and full Navier-Stokes solutions for hypersonic shock-wave/boundary layer interaction, and hydrogen/air supersonic combustion. In addition, other examples of CFD efforts in turbomachinery Navier-Stokes methodology and separated flow modeling are presented. A brief discussion of the 3-tier scientific computing environment is also presented, in which the researcher has access to workstations, mid-size computers, and supercomputers.

ParaBTM: A Parallel Processing Framework for Biomedical Text Mining on Supercomputers.

Science.gov (United States)

Xing, Yuting; Wu, Chengkun; Yang, Xi; Wang, Wei; Zhu, En; Yin, Jianping

2018-04-27

A prevailing way of extracting valuable information from biomedical literature is to apply text mining methods on unstructured texts. However, the massive amount of literature that needs to be analyzed poses a big data challenge to the processing efficiency of text mining. In this paper, we address this challenge by introducing parallel processing on a supercomputer. We developed paraBTM, a runnable framework that enables parallel text mining on the Tianhe-2 supercomputer. It employs a low-cost yet effective load balancing strategy to maximize the efficiency of parallel processing. We evaluated the performance of paraBTM on several datasets, utilizing three types of named entity recognition tasks as demonstration. Results show that, in most cases, the processing efficiency can be greatly improved with parallel processing, and the proposed load balancing strategy is simple and effective. In addition, our framework can be readily applied to other tasks of biomedical text mining besides NER.

Explaining the gap between theoretical peak performance and real performance for supercomputer architectures

International Nuclear Information System (INIS)

Schoenauer, W.; Haefner, H.

1993-01-01

The basic architectures of vector and parallel computers with their properties are presented. Then the memory size and the arithmetic operations in the context of memory bandwidth are discussed. For the exemplary discussion of a single operation micro-measurements of the vector triad for the IBM 3090 VF and the CRAY Y-MP/8 are presented. They reveal the details of the losses for a single operation. Then we analyze the global performance of a whole supercomputer by identifying reduction factors that bring down the theoretical peak performance to the poor real performance. The responsibilities of the manufacturer and of the user for these losses are dicussed. Then the price-performance ratio for different architectures in a snapshot of January 1991 is briefly mentioned. Finally some remarks to a user-friendly architecture for a supercomputer will be made. (orig.)

HPL and STREAM Benchmarks on SANAM Supercomputer

KAUST Repository

Bin Sulaiman, Riman A.

2017-01-01

SANAM supercomputer was jointly built by KACST and FIAS in 2012 ranking second that year in the Green500 list with a power efficiency of 2.3 GFLOPS/W (Rohr et al., 2014). It is a heterogeneous accelerator-based HPC system that has 300 compute nodes. Each node includes two Intel Xeon E5?2650 CPUs, two AMD FirePro S10000 dual GPUs and 128 GiB of main memory. In this work, the seven benchmarks of HPCC were installed and configured to reassess the performance of SANAM, as part of an unpublished master thesis, after it was reassembled in the Kingdom of Saudi Arabia. We present here detailed results of HPL and STREAM benchmarks.

HPL and STREAM Benchmarks on SANAM Supercomputer

KAUST Repository

Bin Sulaiman, Riman A.

2017-03-13

SANAM supercomputer was jointly built by KACST and FIAS in 2012 ranking second that year in the Green500 list with a power efficiency of 2.3 GFLOPS/W (Rohr et al., 2014). It is a heterogeneous accelerator-based HPC system that has 300 compute nodes. Each node includes two Intel Xeon E5?2650 CPUs, two AMD FirePro S10000 dual GPUs and 128 GiB of main memory. In this work, the seven benchmarks of HPCC were installed and configured to reassess the performance of SANAM, as part of an unpublished master thesis, after it was reassembled in the Kingdom of Saudi Arabia. We present here detailed results of HPL and STREAM benchmarks.

An efficient implementation of a backpropagation learning algorithm on quadrics parallel supercomputer

International Nuclear Information System (INIS)

Taraglio, S.; Massaioli, F.

1995-08-01

A parallel implementation of a library to build and train Multi Layer Perceptrons via the Back Propagation algorithm is presented. The target machine is the SIMD massively parallel supercomputer Quadrics. Performance measures are provided on three different machines with different number of processors, for two network examples. A sample source code is given

Supercomputing Centers and Electricity Service Providers

DEFF Research Database (Denmark)

Patki, Tapasya; Bates, Natalie; Ghatikar, Girish

2016-01-01

from a detailed, quantitative survey-based analysis and compare the perspectives of the European grid and SCs to the ones of the United States (US). We then show that contrary to the expectation, SCs in the US are more open toward cooperating and developing demand-management strategies with their ESPs......Supercomputing Centers (SCs) have high and variable power demands, which increase the challenges of the Electricity Service Providers (ESPs) with regards to efficient electricity distribution and reliable grid operation. High penetration of renewable energy generation further exacerbates...... this problem. In order to develop a symbiotic relationship between the SCs and their ESPs and to support effective power management at all levels, it is critical to understand and analyze how the existing relationships were formed and how these are expected to evolve. In this paper, we first present results...

Integration of PanDA workload management system with Titan supercomputer at OLCF

Science.gov (United States)

De, K.; Klimentov, A.; Oleynik, D.; Panitkin, S.; Petrosyan, A.; Schovancova, J.; Vaniachine, A.; Wenaus, T.

2015-12-01

The PanDA (Production and Distributed Analysis) workload management system (WMS) was developed to meet the scale and complexity of LHC distributed computing for the ATLAS experiment. While PanDA currently distributes jobs to more than 100,000 cores at well over 100 Grid sites, the future LHC data taking runs will require more resources than Grid computing can possibly provide. To alleviate these challenges, ATLAS is engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. We will describe a project aimed at integration of PanDA WMS with Titan supercomputer at Oak Ridge Leadership Computing Facility (OLCF). The current approach utilizes a modified PanDA pilot framework for job submission to Titan's batch queues and local data management, with light-weight MPI wrappers to run single threaded workloads in parallel on Titan's multicore worker nodes. It also gives PanDA new capability to collect, in real time, information about unused worker nodes on Titan, which allows precise definition of the size and duration of jobs submitted to Titan according to available free resources. This capability significantly reduces PanDA job wait time while improving Titan's utilization efficiency. This implementation was tested with a variety of Monte-Carlo workloads on Titan and is being tested on several other supercomputing platforms. Notice: This manuscript has been authored, by employees of Brookhaven Science Associates, LLC under Contract No. DE-AC02-98CH10886 with the U.S. Department of Energy. The publisher by accepting the manuscript for publication acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes.

Building more powerful less expensive supercomputers using Processing-In-Memory (PIM) LDRD final report.

Energy Technology Data Exchange (ETDEWEB)

Murphy, Richard C.

2009-09-01

This report details the accomplishments of the 'Building More Powerful Less Expensive Supercomputers Using Processing-In-Memory (PIM)' LDRD ('PIM LDRD', number 105809) for FY07-FY09. Latency dominates all levels of supercomputer design. Within a node, increasing memory latency, relative to processor cycle time, limits CPU performance. Between nodes, the same increase in relative latency impacts scalability. Processing-In-Memory (PIM) is an architecture that directly addresses this problem using enhanced chip fabrication technology and machine organization. PIMs combine high-speed logic and dense, low-latency, high-bandwidth DRAM, and lightweight threads that tolerate latency by performing useful work during memory transactions. This work examines the potential of PIM-based architectures to support mission critical Sandia applications and an emerging class of more data intensive informatics applications. This work has resulted in a stronger architecture/implementation collaboration between 1400 and 1700. Additionally, key technology components have impacted vendor roadmaps, and we are in the process of pursuing these new collaborations. This work has the potential to impact future supercomputer design and construction, reducing power and increasing performance. This final report is organized as follow: this summary chapter discusses the impact of the project (Section 1), provides an enumeration of publications and other public discussion of the work (Section 1), and concludes with a discussion of future work and impact from the project (Section 1). The appendix contains reprints of the refereed publications resulting from this work.

Supercomputers and the mathematical modeling of high complexity problems

International Nuclear Information System (INIS)

Belotserkovskii, Oleg M

2010-01-01

This paper is a review of many works carried out by members of our scientific school in past years. The general principles of constructing numerical algorithms for high-performance computers are described. Several techniques are highlighted and these are based on the method of splitting with respect to physical processes and are widely used in computing nonlinear multidimensional processes in fluid dynamics, in studies of turbulence and hydrodynamic instabilities and in medicine and other natural sciences. The advances and developments related to the new generation of high-performance supercomputing in Russia are presented.

The Pawsey Supercomputer geothermal cooling project

Science.gov (United States)

Regenauer-Lieb, K.; Horowitz, F.; Western Australian Geothermal Centre Of Excellence, T.

2010-12-01

The Australian Government has funded the Pawsey supercomputer in Perth, Western Australia, providing computational infrastructure intended to support the future operations of the Australian Square Kilometre Array radiotelescope and to boost next-generation computational geosciences in Australia. Supplementary funds have been directed to the development of a geothermal exploration well to research the potential for direct heat use applications at the Pawsey Centre site. Cooling the Pawsey supercomputer may be achieved by geothermal heat exchange rather than by conventional electrical power cooling, thus reducing the carbon footprint of the Pawsey Centre and demonstrating an innovative green technology that is widely applicable in industry and urban centres across the world. The exploration well is scheduled to be completed in 2013, with drilling due to commence in the third quarter of 2011. One year is allocated to finalizing the design of the exploration, monitoring and research well. Success in the geothermal exploration and research program will result in an industrial-scale geothermal cooling facility at the Pawsey Centre, and will provide a world-class student training environment in geothermal energy systems. A similar system is partially funded and in advanced planning to provide base-load air-conditioning for the main campus of the University of Western Australia. Both systems are expected to draw ~80-95 degrees C water from aquifers lying between 2000 and 3000 meters depth from naturally permeable rocks of the Perth sedimentary basin. The geothermal water will be run through absorption chilling devices, which only require heat (as opposed to mechanical work) to power a chilled water stream adequate to meet the cooling requirements. Once the heat has been removed from the geothermal water, licensing issues require the water to be re-injected back into the aquifer system. These systems are intended to demonstrate the feasibility of powering large-scale air

Heat dissipation computations of a HVDC ground electrode using a supercomputer

International Nuclear Information System (INIS)

Greiss, H.; Mukhedkar, D.; Lagace, P.J.

1990-01-01

This paper reports on the temperature, of soil surrounding a High Voltage Direct Current (HVDC) toroidal ground electrode of practical dimensions, in both homogeneous and non-homogeneous soils that was computed at incremental points in time using finite difference methods on a supercomputer. Curves of the response were computed and plotted at several locations within the soil in the vicinity of the ground electrode for various values of the soil parameters

Argonne National Lab deploys Force10 networks' massively dense ethernet switch for supercomputing cluster

CERN Multimedia

2003-01-01

"Force10 Networks, Inc. today announced that Argonne National Laboratory (Argonne, IL) has successfully deployed Force10 E-Series switch/routers to connect to the TeraGrid, the world's largest supercomputing grid, sponsored by the National Science Foundation (NSF)" (1/2 page).

A supercomputing application for reactors core design and optimization

International Nuclear Information System (INIS)

Hourcade, Edouard; Gaudier, Fabrice; Arnaud, Gilles; Funtowiez, David; Ammar, Karim

2010-01-01

Advanced nuclear reactor designs are often intuition-driven processes where designers first develop or use simplified simulation tools for each physical phenomenon involved. Through the project development, complexity in each discipline increases and implementation of chaining/coupling capabilities adapted to supercomputing optimization process are often postponed to a further step so that task gets increasingly challenging. In the context of renewal in reactor designs, project of first realization are often run in parallel with advanced design although very dependant on final options. As a consequence, the development of tools to globally assess/optimize reactor core features, with the on-going design methods accuracy, is needed. This should be possible within reasonable simulation time and without advanced computer skills needed at project management scale. Also, these tools should be ready to easily cope with modeling progresses in each discipline through project life-time. An early stage development of multi-physics package adapted to supercomputing is presented. The URANIE platform, developed at CEA and based on the Data Analysis Framework ROOT, is very well adapted to this approach. It allows diversified sampling techniques (SRS, LHS, qMC), fitting tools (neuronal networks...) and optimization techniques (genetic algorithm). Also data-base management and visualization are made very easy. In this paper, we'll present the various implementing steps of this core physics tool where neutronics, thermo-hydraulics, and fuel mechanics codes are run simultaneously. A relevant example of optimization of nuclear reactor safety characteristics will be presented. Also, flexibility of URANIE tool will be illustrated with the presentation of several approaches to improve Pareto front quality. (author)

Performance Evaluation of Supercomputers using HPCC and IMB Benchmarks

Science.gov (United States)

Saini, Subhash; Ciotti, Robert; Gunney, Brian T. N.; Spelce, Thomas E.; Koniges, Alice; Dossa, Don; Adamidis, Panagiotis; Rabenseifner, Rolf; Tiyyagura, Sunil R.; Mueller, Matthias;

An Interface for Biomedical Big Data Processing on the Tianhe-2 Supercomputer.

Science.gov (United States)

Yang, Xi; Wu, Chengkun; Lu, Kai; Fang, Lin; Zhang, Yong; Li, Shengkang; Guo, Guixin; Du, YunFei

2017-12-01

Big data, cloud computing, and high-performance computing (HPC) are at the verge of convergence. Cloud computing is already playing an active part in big data processing with the help of big data frameworks like Hadoop and Spark. The recent upsurge of high-performance computing in China provides extra possibilities and capacity to address the challenges associated with big data. In this paper, we propose Orion-a big data interface on the Tianhe-2 supercomputer-to enable big data applications to run on Tianhe-2 via a single command or a shell script. Orion supports multiple users, and each user can launch multiple tasks. It minimizes the effort needed to initiate big data applications on the Tianhe-2 supercomputer via automated configuration. Orion follows the "allocate-when-needed" paradigm, and it avoids the idle occupation of computational resources. We tested the utility and performance of Orion using a big genomic dataset and achieved a satisfactory performance on Tianhe-2 with very few modifications to existing applications that were implemented in Hadoop/Spark. In summary, Orion provides a practical and economical interface for big data processing on Tianhe-2.

Supercomputations and big-data analysis in strong-field ultrafast optical physics: filamentation of high-peak-power ultrashort laser pulses

Science.gov (United States)

Voronin, A. A.; Panchenko, V. Ya; Zheltikov, A. M.

2016-06-01

High-intensity ultrashort laser pulses propagating in gas media or in condensed matter undergo complex nonlinear spatiotemporal evolution where temporal transformations of optical field waveforms are strongly coupled to an intricate beam dynamics and ultrafast field-induced ionization processes. At the level of laser peak powers orders of magnitude above the critical power of self-focusing, the beam exhibits modulation instabilities, producing random field hot spots and breaking up into multiple noise-seeded filaments. This problem is described by a (3  +  1)-dimensional nonlinear field evolution equation, which needs to be solved jointly with the equation for ultrafast ionization of a medium. Analysis of this problem, which is equivalent to solving a billion-dimensional evolution problem, is only possible by means of supercomputer simulations augmented with coordinated big-data processing of large volumes of information acquired through theory-guiding experiments and supercomputations. Here, we review the main challenges of supercomputations and big-data processing encountered in strong-field ultrafast optical physics and discuss strategies to confront these challenges.

Quantum Hamiltonian Physics with Supercomputers

International Nuclear Information System (INIS)

Vary, James P.

2014-01-01

The vision of solving the nuclear many-body problem in a Hamiltonian framework with fundamental interactions tied to QCD via Chiral Perturbation Theory is gaining support. The goals are to preserve the predictive power of the underlying theory, to test fundamental symmetries with the nucleus as laboratory and to develop new understandings of the full range of complex quantum phenomena. Advances in theoretical frameworks (renormalization and many-body methods) as well as in computational resources (new algorithms and leadership-class parallel computers) signal a new generation of theory and simulations that will yield profound insights into the origins of nuclear shell structure, collective phenomena and complex reaction dynamics. Fundamental discovery opportunities also exist in such areas as physics beyond the Standard Model of Elementary Particles, the transition between hadronic and quark–gluon dominated dynamics in nuclei and signals that characterize dark matter. I will review some recent achievements and present ambitious consensus plans along with their challenges for a coming decade of research that will build new links between theory, simulations and experiment. Opportunities for graduate students to embark upon careers in the fast developing field of supercomputer simulations is also discussed

Quantum Hamiltonian Physics with Supercomputers

Energy Technology Data Exchange (ETDEWEB)

Vary, James P.

2014-06-15

The vision of solving the nuclear many-body problem in a Hamiltonian framework with fundamental interactions tied to QCD via Chiral Perturbation Theory is gaining support. The goals are to preserve the predictive power of the underlying theory, to test fundamental symmetries with the nucleus as laboratory and to develop new understandings of the full range of complex quantum phenomena. Advances in theoretical frameworks (renormalization and many-body methods) as well as in computational resources (new algorithms and leadership-class parallel computers) signal a new generation of theory and simulations that will yield profound insights into the origins of nuclear shell structure, collective phenomena and complex reaction dynamics. Fundamental discovery opportunities also exist in such areas as physics beyond the Standard Model of Elementary Particles, the transition between hadronic and quark–gluon dominated dynamics in nuclei and signals that characterize dark matter. I will review some recent achievements and present ambitious consensus plans along with their challenges for a coming decade of research that will build new links between theory, simulations and experiment. Opportunities for graduate students to embark upon careers in the fast developing field of supercomputer simulations is also discussed.

Coherent 40 Gb/s SP-16QAM and 80 Gb/s PDM-16QAM in an Optimal Supercomputer Optical Switch Fabric

DEFF Research Database (Denmark)

Karinou, Fotini; Borkowski, Robert; Zibar, Darko

2013-01-01

We demonstrate, for the first time, the feasibility of using 40 Gb/s SP-16QAM and 80 Gb/s PDM-16QAM in an optimized cell switching supercomputer optical interconnect architecture based on semiconductor optical amplifiers as ON/OFF gates.......We demonstrate, for the first time, the feasibility of using 40 Gb/s SP-16QAM and 80 Gb/s PDM-16QAM in an optimized cell switching supercomputer optical interconnect architecture based on semiconductor optical amplifiers as ON/OFF gates....

Cooperative visualization and simulation in a supercomputer environment

International Nuclear Information System (INIS)

Ruehle, R.; Lang, U.; Wierse, A.

1993-01-01

The article takes a closer look on the requirements being imposed by the idea to integrate all the components into a homogeneous software environment. To this end several methods for the distribtuion of applications in dependence of certain problem types are discussed. The currently available methods at the University of Stuttgart Computer Center for the distribution of applications are further explained. Finally the aims and characteristics of a European sponsored project, called PAGEIN, are explained, which fits perfectly into the line of developments at RUS. The aim of the project is to experiment with future cooperative working modes of aerospace scientists in a high speed distributed supercomputing environment. Project results will have an impact on the development of real future scientific application environments. (orig./DG)

Feynman diagrams sampling for quantum field theories on the QPACE 2 supercomputer

Energy Technology Data Exchange (ETDEWEB)

Rappl, Florian

2016-08-01

This work discusses the application of Feynman diagram sampling in quantum field theories. The method uses a computer simulation to sample the diagrammatic space obtained in a series expansion. For running large physical simulations powerful computers are obligatory, effectively splitting the thesis in two parts. The first part deals with the method of Feynman diagram sampling. Here the theoretical background of the method itself is discussed. Additionally, important statistical concepts and the theory of the strong force, quantum chromodynamics, are introduced. This sets the context of the simulations. We create and evaluate a variety of models to estimate the applicability of diagrammatic methods. The method is then applied to sample the perturbative expansion of the vertex correction. In the end we obtain the value for the anomalous magnetic moment of the electron. The second part looks at the QPACE 2 supercomputer. This includes a short introduction to supercomputers in general, as well as a closer look at the architecture and the cooling system of QPACE 2. Guiding benchmarks of the InfiniBand network are presented. At the core of this part, a collection of best practices and useful programming concepts are outlined, which enables the development of efficient, yet easily portable, applications for the QPACE 2 system.

Use of high performance networks and supercomputers for real-time flight simulation

Science.gov (United States)

Cleveland, Jeff I., II

1993-01-01

In order to meet the stringent time-critical requirements for real-time man-in-the-loop flight simulation, computer processing operations must be consistent in processing time and be completed in as short a time as possible. These operations include simulation mathematical model computation and data input/output to the simulators. In 1986, in response to increased demands for flight simulation performance, NASA's Langley Research Center (LaRC), working with the contractor, developed extensions to the Computer Automated Measurement and Control (CAMAC) technology which resulted in a factor of ten increase in the effective bandwidth and reduced latency of modules necessary for simulator communication. This technology extension is being used by more than 80 leading technological developers in the United States, Canada, and Europe. Included among the commercial applications are nuclear process control, power grid analysis, process monitoring, real-time simulation, and radar data acquisition. Personnel at LaRC are completing the development of the use of supercomputers for mathematical model computation to support real-time flight simulation. This includes the development of a real-time operating system and development of specialized software and hardware for the simulator network. This paper describes the data acquisition technology and the development of supercomputing for flight simulation.

Federal Market Information Technology in the Post Flash Crash Era: Roles for Supercomputing

Energy Technology Data Exchange (ETDEWEB)

Bethel, E. Wes; Leinweber, David; Ruebel, Oliver; Wu, Kesheng

2011-09-16

This paper describes collaborative work between active traders, regulators, economists, and supercomputing researchers to replicate and extend investigations of the Flash Crash and other market anomalies in a National Laboratory HPC environment. Our work suggests that supercomputing tools and methods will be valuable to market regulators in achieving the goal of market safety, stability, and security. Research results using high frequency data and analytics are described, and directions for future development are discussed. Currently the key mechanism for preventing catastrophic market action are “circuit breakers.” We believe a more graduated approach, similar to the “yellow light” approach in motorsports to slow down traffic, might be a better way to achieve the same goal. To enable this objective, we study a number of indicators that could foresee hazards in market conditions and explore options to confirm such predictions. Our tests confirm that Volume Synchronized Probability of Informed Trading (VPIN) and a version of volume Herfindahl-Hirschman Index (HHI) for measuring market fragmentation can indeed give strong signals ahead of the Flash Crash event on May 6 2010. This is a preliminary step toward a full-fledged early-warning system for unusual market conditions.

A fast random number generator for the Intel Paragon supercomputer

Science.gov (United States)

Gutbrod, F.

1995-06-01

A pseudo-random number generator is presented which makes optimal use of the architecture of the i860-microprocessor and which is expected to have a very long period. It is therefore a good candidate for use on the parallel supercomputer Paragon XP. In the assembler version, it needs 6.4 cycles for a real∗4 random number. There is a FORTRAN routine which yields identical numbers up to rare and minor rounding discrepancies, and it needs 28 cycles. The FORTRAN performance on other microprocessors is somewhat better. Arguments for the quality of the generator and some numerical tests are given.

Development of a Cloud Resolving Model for Heterogeneous Supercomputers

Science.gov (United States)

Sreepathi, S.; Norman, M. R.; Pal, A.; Hannah, W.; Ponder, C.

2017-12-01

A cloud resolving climate model is needed to reduce major systematic errors in climate simulations due to structural uncertainty in numerical treatments of convection - such as convective storm systems. This research describes the porting effort to enable SAM (System for Atmosphere Modeling) cloud resolving model on heterogeneous supercomputers using GPUs (Graphical Processing Units). We have isolated a standalone configuration of SAM that is targeted to be integrated into the DOE ACME (Accelerated Climate Modeling for Energy) Earth System model. We have identified key computational kernels from the model and offloaded them to a GPU using the OpenACC programming model. Furthermore, we are investigating various optimization strategies intended to enhance GPU utilization including loop fusion/fission, coalesced data access and loop refactoring to a higher abstraction level. We will present early performance results, lessons learned as well as optimization strategies. The computational platform used in this study is the Summitdev system, an early testbed that is one generation removed from Summit, the next leadership class supercomputer at Oak Ridge National Laboratory. The system contains 54 nodes wherein each node has 2 IBM POWER8 CPUs and 4 NVIDIA Tesla P100 GPUs. This work is part of a larger project, ACME-MMF component of the U.S. Department of Energy(DOE) Exascale Computing Project. The ACME-MMF approach addresses structural uncertainty in cloud processes by replacing traditional parameterizations with cloud resolving "superparameterization" within each grid cell of global climate model. Super-parameterization dramatically increases arithmetic intensity, making the MMF approach an ideal strategy to achieve good performance on emerging exascale computing architectures. The goal of the project is to integrate superparameterization into ACME, and explore its full potential to scientifically and computationally advance climate simulation and prediction.

Frequently updated noise threat maps created with use of supercomputing grid

Directory of Open Access Journals (Sweden)

Szczodrak Maciej

2014-09-01

Full Text Available An innovative supercomputing grid services devoted to noise threat evaluation were presented. The services described in this paper concern two issues, first is related to the noise mapping, while the second one focuses on assessment of the noise dose and its influence on the human hearing system. The discussed serviceswere developed within the PL-Grid Plus Infrastructure which accumulates Polish academic supercomputer centers. Selected experimental results achieved by the usage of the services proposed were presented. The assessment of the environmental noise threats includes creation of the noise maps using either ofline or online data, acquired through a grid of the monitoring stations. A concept of estimation of the source model parameters based on the measured sound level for the purpose of creating frequently updated noise maps was presented. Connecting the noise mapping grid service with a distributed sensor network enables to automatically update noise maps for a specified time period. Moreover, a unique attribute of the developed software is the estimation of the auditory effects evoked by the exposure to noise. The estimation method uses a modified psychoacoustic model of hearing and is based on the calculated noise level values and on the given exposure period. Potential use scenarios of the grid services for research or educational purpose were introduced. Presentation of the results of predicted hearing threshold shift caused by exposure to excessive noise can raise the public awareness of the noise threats.

Performance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore supercomputers

KAUST Repository

Wu, Xingfu; Taylor, Valerie

2011-01-01

The NAS Parallel Benchmarks (NPB) are well-known applications with the fixed algorithms for evaluating parallel systems and tools. Multicore supercomputers provide a natural programming paradigm for hybrid programs, whereby OpenMP can be used with the data sharing with the multicores that comprise a node and MPI can be used with the communication between nodes. In this paper, we use SP and BT benchmarks of MPI NPB 3.3 as a basis for a comparative approach to implement hybrid MPI/OpenMP versions of SP and BT. In particular, we can compare the performance of the hybrid SP and BT with the MPI counterparts on large-scale multicore supercomputers. Our performance results indicate that the hybrid SP outperforms the MPI SP by up to 20.76%, and the hybrid BT outperforms the MPI BT by up to 8.58% on up to 10,000 cores on BlueGene/P at Argonne National Laboratory and Jaguar (Cray XT4/5) at Oak Ridge National Laboratory. We also use performance tools and MPI trace libraries available on these supercomputers to further investigate the performance characteristics of the hybrid SP and BT.

Performance characteristics of hybrid MPI/OpenMP implementations of NAS parallel benchmarks SP and BT on large-scale multicore supercomputers

KAUST Repository

Wu, Xingfu

2011-03-29

The NAS Parallel Benchmarks (NPB) are well-known applications with the fixed algorithms for evaluating parallel systems and tools. Multicore supercomputers provide a natural programming paradigm for hybrid programs, whereby OpenMP can be used with the data sharing with the multicores that comprise a node and MPI can be used with the communication between nodes. In this paper, we use SP and BT benchmarks of MPI NPB 3.3 as a basis for a comparative approach to implement hybrid MPI/OpenMP versions of SP and BT. In particular, we can compare the performance of the hybrid SP and BT with the MPI counterparts on large-scale multicore supercomputers. Our performance results indicate that the hybrid SP outperforms the MPI SP by up to 20.76%, and the hybrid BT outperforms the MPI BT by up to 8.58% on up to 10,000 cores on BlueGene/P at Argonne National Laboratory and Jaguar (Cray XT4/5) at Oak Ridge National Laboratory. We also use performance tools and MPI trace libraries available on these supercomputers to further investigate the performance characteristics of the hybrid SP and BT.

Simulation of x-rays in refractive structure by the Monte Carlo method using the supercomputer SKIF

International Nuclear Information System (INIS)

Yaskevich, Yu.R.; Kravchenko, O.I.; Soroka, I.I.; Chembrovskij, A.G.; Kolesnik, A.S.; Serikova, N.V.; Petrov, P.V.; Kol'chevskij, N.N.

2013-01-01

Software 'Xray-SKIF' for the simulation of the X-rays in refractive structures by the Monte-Carlo method using the supercomputer SKIF BSU are developed. The program generates a large number of rays propagated from a source to the refractive structure. The ray trajectory under assumption of geometrical optics is calculated. Absorption is calculated for each ray inside of refractive structure. Dynamic arrays are used for results of calculation rays parameters, its restore the X-ray field distributions very fast at different position of detector. It was found that increasing the number of processors leads to proportional decreasing of calculation time: simulation of 10 8 X-rays using supercomputer with the number of processors from 1 to 30 run-times equal 3 hours and 6 minutes, respectively. 10 9 X-rays are calculated by software 'Xray-SKIF' which allows to reconstruct the X-ray field after refractive structure with a special resolution of 1 micron. (authors)

Communication Characterization and Optimization of Applications Using Topology-Aware Task Mapping on Large Supercomputers

Energy Technology Data Exchange (ETDEWEB)

Sreepathi, Sarat [ORNL; D' Azevedo, Eduardo [ORNL; Philip, Bobby [ORNL; Worley, Patrick H [ORNL

2016-01-01

On large supercomputers, the job scheduling systems may assign a non-contiguous node allocation for user applications depending on available resources. With parallel applications using MPI (Message Passing Interface), the default process ordering does not take into account the actual physical node layout available to the application. This contributes to non-locality in terms of physical network topology and impacts communication performance of the application. In order to mitigate such performance penalties, this work describes techniques to identify suitable task mapping that takes the layout of the allocated nodes as well as the application's communication behavior into account. During the first phase of this research, we instrumented and collected performance data to characterize communication behavior of critical US DOE (United States - Department of Energy) applications using an augmented version of the mpiP tool. Subsequently, we developed several reordering methods (spectral bisection, neighbor join tree etc.) to combine node layout and application communication data for optimized task placement. We developed a tool called mpiAproxy to facilitate detailed evaluation of the various reordering algorithms without requiring full application executions. This work presents a comprehensive performance evaluation (14,000 experiments) of the various task mapping techniques in lowering communication costs on Titan, the leadership class supercomputer at Oak Ridge National Laboratory.

Plasma turbulence calculations on supercomputers

International Nuclear Information System (INIS)

Carreras, B.A.; Charlton, L.A.; Dominguez, N.; Drake, J.B.; Garcia, L.; Leboeuf, J.N.; Lee, D.K.; Lynch, V.E.; Sidikman, K.

1991-01-01

Although the single-particle picture of magnetic confinement is helpful in understanding some basic physics of plasma confinement, it does not give a full description. Collective effects dominate plasma behavior. Any analysis of plasma confinement requires a self-consistent treatment of the particles and fields. The general picture is further complicated because the plasma, in general, is turbulent. The study of fluid turbulence is a rather complex field by itself. In addition to the difficulties of classical fluid turbulence, plasma turbulence studies face the problems caused by the induced magnetic turbulence, which couples field by itself. In addition to the difficulties of classical fluid turbulence, plasma turbulence studies face the problems caused by the induced magnetic turbulence, which couples back to the fluid. Since the fluid is not a perfect conductor, this turbulence can lead to changes in the topology of the magnetic field structure, causing the magnetic field lines to wander radially. Because the plasma fluid flows along field lines, they carry the particles with them, and this enhances the losses caused by collisions. The changes in topology are critical for the plasma confinement. The study of plasma turbulence and the concomitant transport is a challenging problem. Because of the importance of solving the plasma turbulence problem for controlled thermonuclear research, the high complexity of the problem, and the necessity of attacking the problem with supercomputers, the study of plasma turbulence in magnetic confinement devices is a Grand Challenge problem

Reactive flow simulations in complex geometries with high-performance supercomputing

International Nuclear Information System (INIS)

Rehm, W.; Gerndt, M.; Jahn, W.; Vogelsang, R.; Binninger, B.; Herrmann, M.; Olivier, H.; Weber, M.

2000-01-01

In this paper, we report on a modern field code cluster consisting of state-of-the-art reactive Navier-Stokes- and reactive Euler solvers that has been developed on vector- and parallel supercomputers at the research center Juelich. This field code cluster is used for hydrogen safety analyses of technical systems, for example, in the field of nuclear reactor safety and conventional hydrogen demonstration plants with fuel cells. Emphasis is put on the assessment of combustion loads, which could result from slow, fast or rapid flames, including transition from deflagration to detonation. As a sample of proof tests, the special tools have been tested for specific tasks, based on the comparison of experimental and numerical results, which are in reasonable agreement. (author)

Storage-Intensive Supercomputing Benchmark Study

Energy Technology Data Exchange (ETDEWEB)

Cohen, J; Dossa, D; Gokhale, M; Hysom, D; May, J; Pearce, R; Yoo, A

2007-10-30

Critical data science applications requiring frequent access to storage perform poorly on today's computing architectures. This project addresses efficient computation of data-intensive problems in national security and basic science by exploring, advancing, and applying a new form of computing called storage-intensive supercomputing (SISC). Our goal is to enable applications that simply cannot run on current systems, and, for a broad range of data-intensive problems, to deliver an order of magnitude improvement in price/performance over today's data-intensive architectures. This technical report documents much of the work done under LDRD 07-ERD-063 Storage Intensive Supercomputing during the period 05/07-09/07. The following chapters describe: (1) a new file I/O monitoring tool iotrace developed to capture the dynamic I/O profiles of Linux processes; (2) an out-of-core graph benchmark for level-set expansion of scale-free graphs; (3) an entity extraction benchmark consisting of a pipeline of eight components; and (4) an image resampling benchmark drawn from the SWarp program in the LSST data processing pipeline. The performance of the graph and entity extraction benchmarks was measured in three different scenarios: data sets residing on the NFS file server and accessed over the network; data sets stored on local disk; and data sets stored on the Fusion I/O parallel NAND Flash array. The image resampling benchmark compared performance of software-only to GPU-accelerated. In addition to the work reported here, an additional text processing application was developed that used an FPGA to accelerate n-gram profiling for language classification. The n-gram application will be presented at SC07 at the High Performance Reconfigurable Computing Technologies and Applications Workshop. The graph and entity extraction benchmarks were run on a Supermicro server housing the NAND Flash 40GB parallel disk array, the Fusion-io. The Fusion system specs are as follows

Parallel Multivariate Spatio-Temporal Clustering of Large Ecological Datasets on Hybrid Supercomputers

Energy Technology Data Exchange (ETDEWEB)

Sreepathi, Sarat [ORNL; Kumar, Jitendra [ORNL; Mills, Richard T. [Argonne National Laboratory; Hoffman, Forrest M. [ORNL; Sripathi, Vamsi [Intel Corporation; Hargrove, William Walter [United States Department of Agriculture (USDA), United States Forest Service (USFS)

2017-09-01

A proliferation of data from vast networks of remote sensing platforms (satellites, unmanned aircraft systems (UAS), airborne etc.), observational facilities (meteorological, eddy covariance etc.), state-of-the-art sensors, and simulation models offer unprecedented opportunities for scientific discovery. Unsupervised classification is a widely applied data mining approach to derive insights from such data. However, classification of very large data sets is a complex computational problem that requires efficient numerical algorithms and implementations on high performance computing (HPC) platforms. Additionally, increasing power, space, cooling and efficiency requirements has led to the deployment of hybrid supercomputing platforms with complex architectures and memory hierarchies like the Titan system at Oak Ridge National Laboratory. The advent of such accelerated computing architectures offers new challenges and opportunities for big data analytics in general and specifically, large scale cluster analysis in our case. Although there is an existing body of work on parallel cluster analysis, those approaches do not fully meet the needs imposed by the nature and size of our large data sets. Moreover, they had scaling limitations and were mostly limited to traditional distributed memory computing platforms. We present a parallel Multivariate Spatio-Temporal Clustering (MSTC) technique based on k-means cluster analysis that can target hybrid supercomputers like Titan. We developed a hybrid MPI, CUDA and OpenACC implementation that can utilize both CPU and GPU resources on computational nodes. We describe performance results on Titan that demonstrate the scalability and efficacy of our approach in processing large ecological data sets.

High temporal resolution mapping of seismic noise sources using heterogeneous supercomputers

Science.gov (United States)

Gokhberg, Alexey; Ermert, Laura; Paitz, Patrick; Fichtner, Andreas

2017-04-01

Time- and space-dependent distribution of seismic noise sources is becoming a key ingredient of modern real-time monitoring of various geo-systems. Significant interest in seismic noise source maps with high temporal resolution (days) is expected to come from a number of domains, including natural resources exploration, analysis of active earthquake fault zones and volcanoes, as well as geothermal and hydrocarbon reservoir monitoring. Currently, knowledge of noise sources is insufficient for high-resolution subsurface monitoring applications. Near-real-time seismic data, as well as advanced imaging methods to constrain seismic noise sources have recently become available. These methods are based on the massive cross-correlation of seismic noise records from all available seismic stations in the region of interest and are therefore very computationally intensive. Heterogeneous massively parallel supercomputing systems introduced in the recent years combine conventional multi-core CPU with GPU accelerators and provide an opportunity for manifold increase and computing performance. Therefore, these systems represent an efficient platform for implementation of a noise source mapping solution. We present the first results of an ongoing research project conducted in collaboration with the Swiss National Supercomputing Centre (CSCS). The project aims at building a service that provides seismic noise source maps for Central Europe with high temporal resolution (days to few weeks depending on frequency and data availability). The service is hosted on the CSCS computing infrastructure; all computationally intensive processing is performed on the massively parallel heterogeneous supercomputer "Piz Daint". The solution architecture is based on the Application-as-a-Service concept in order to provide the interested external researchers the regular access to the noise source maps. The solution architecture includes the following sub-systems: (1) data acquisition responsible for

Harnessing Petaflop-Scale Multi-Core Supercomputing for Problems in Space Science

Science.gov (United States)

Albright, B. J.; Yin, L.; Bowers, K. J.; Daughton, W.; Bergen, B.; Kwan, T. J.

2008-12-01

The particle-in-cell kinetic plasma code VPIC has been migrated successfully to the world's fastest supercomputer, Roadrunner, a hybrid multi-core platform built by IBM for the Los Alamos National Laboratory. How this was achieved will be described and examples of state-of-the-art calculations in space science, in particular, the study of magnetic reconnection, will be presented. With VPIC on Roadrunner, we have performed, for the first time, plasma PIC calculations with over one trillion particles, >100× larger than calculations considered "heroic" by community standards. This allows examination of physics at unprecedented scale and fidelity. Roadrunner is an example of an emerging paradigm in supercomputing: the trend toward multi-core systems with deep hierarchies and where memory bandwidth optimization is vital to achieving high performance. Getting VPIC to perform well on such systems is a formidable challenge: the core algorithm is memory bandwidth limited with low compute-to-data ratio and requires random access to memory in its inner loop. That we were able to get VPIC to perform and scale well, achieving >0.374 Pflop/s and linear weak scaling on real physics problems on up to the full 12240-core Roadrunner machine, bodes well for harnessing these machines for our community's needs in the future. Many of the design considerations encountered commute to other multi-core and accelerated (e.g., via GPU) platforms and we modified VPIC with flexibility in mind. These will be summarized and strategies for how one might adapt a code for such platforms will be shared. Work performed under the auspices of the U.S. DOE by the LANS LLC Los Alamos National Laboratory. Dr. Bowers is a LANL Guest Scientist; he is presently at D. E. Shaw Research LLC, 120 W 45th Street, 39th Floor, New York, NY 10036.

Integration of Titan supercomputer at OLCF with ATLAS Production System

Science.gov (United States)

Barreiro Megino, F.; De, K.; Jha, S.; Klimentov, A.; Maeno, T.; Nilsson, P.; Oleynik, D.; Padolski, S.; Panitkin, S.; Wells, J.; Wenaus, T.; ATLAS Collaboration

2017-10-01

The PanDA (Production and Distributed Analysis) workload management system was developed to meet the scale and complexity of distributed computing for the ATLAS experiment. PanDA managed resources are distributed worldwide, on hundreds of computing sites, with thousands of physicists accessing hundreds of Petabytes of data and the rate of data processing already exceeds Exabyte per year. While PanDA currently uses more than 200,000 cores at well over 100 Grid sites, future LHC data taking runs will require more resources than Grid computing can possibly provide. Additional computing and storage resources are required. Therefore ATLAS is engaged in an ambitious program to expand the current computing model to include additional resources such as the opportunistic use of supercomputers. In this paper we will describe a project aimed at integration of ATLAS Production System with Titan supercomputer at Oak Ridge Leadership Computing Facility (OLCF). Current approach utilizes modified PanDA Pilot framework for job submission to Titan’s batch queues and local data management, with lightweight MPI wrappers to run single node workloads in parallel on Titan’s multi-core worker nodes. It provides for running of standard ATLAS production jobs on unused resources (backfill) on Titan. The system already allowed ATLAS to collect on Titan millions of core-hours per month, execute hundreds of thousands jobs, while simultaneously improving Titans utilization efficiency. We will discuss the details of the implementation, current experience with running the system, as well as future plans aimed at improvements in scalability and efficiency. Notice: This manuscript has been authored, by employees of Brookhaven Science Associates, LLC under Contract No. DE-AC02-98CH10886 with the U.S. Department of Energy. The publisher by accepting the manuscript for publication acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to

Use of QUADRICS supercomputer as embedded simulator in emergency management systems

International Nuclear Information System (INIS)

Bove, R.; Di Costanzo, G.; Ziparo, A.

1996-07-01

The experience related to the implementation of a MRBT, atmospheric spreading model with a short duration releasing, are reported. This model was implemented on a QUADRICS-Q1 supercomputer. First is reported a description of the MRBT model. It is an analytical model to study the speadings of light gases realised in the atmosphere cause incidental releasing. The solution of diffusion equation is Gaussian like. It yield the concentration of pollutant substance released. The concentration is function of space and time. Thus the QUADRICS architecture is introduced. And the implementation of the model is described. At the end it will be consider the integration of the QUADRICS-based model as simulator in a emergency management system

MILC Code Performance on High End CPU and GPU Supercomputer Clusters

Science.gov (United States)

DeTar, Carleton; Gottlieb, Steven; Li, Ruizi; Toussaint, Doug

2018-03-01

With recent developments in parallel supercomputing architecture, many core, multi-core, and GPU processors are now commonplace, resulting in more levels of parallelism, memory hierarchy, and programming complexity. It has been necessary to adapt the MILC code to these new processors starting with NVIDIA GPUs, and more recently, the Intel Xeon Phi processors. We report on our efforts to port and optimize our code for the Intel Knights Landing architecture. We consider performance of the MILC code with MPI and OpenMP, and optimizations with QOPQDP and QPhiX. For the latter approach, we concentrate on the staggered conjugate gradient and gauge force. We also consider performance on recent NVIDIA GPUs using the QUDA library.

MILC Code Performance on High End CPU and GPU Supercomputer Clusters

Directory of Open Access Journals (Sweden)

DeTar Carleton

2018-01-01

Full Text Available With recent developments in parallel supercomputing architecture, many core, multi-core, and GPU processors are now commonplace, resulting in more levels of parallelism, memory hierarchy, and programming complexity. It has been necessary to adapt the MILC code to these new processors starting with NVIDIA GPUs, and more recently, the Intel Xeon Phi processors. We report on our efforts to port and optimize our code for the Intel Knights Landing architecture. We consider performance of the MILC code with MPI and OpenMP, and optimizations with QOPQDP and QPhiX. For the latter approach, we concentrate on the staggered conjugate gradient and gauge force. We also consider performance on recent NVIDIA GPUs using the QUDA library.

SUPERCOMPUTER SIMULATION OF CRITICAL PHENOMENA IN COMPLEX SOCIAL SYSTEMS

Directory of Open Access Journals (Sweden)

Petrus M.A. Sloot

2014-09-01

Full Text Available The paper describes a problem of computer simulation of critical phenomena in complex social systems on a petascale computing systems in frames of complex networks approach. The three-layer system of nested models of complex networks is proposed including aggregated analytical model to identify critical phenomena, detailed model of individualized network dynamics and model to adjust a topological structure of a complex network. The scalable parallel algorithm covering all layers of complex networks simulation is proposed. Performance of the algorithm is studied on different supercomputing systems. The issues of software and information infrastructure of complex networks simulation are discussed including organization of distributed calculations, crawling the data in social networks and results visualization. The applications of developed methods and technologies are considered including simulation of criminal networks disruption, fast rumors spreading in social networks, evolution of financial networks and epidemics spreading.

Lectures in Supercomputational Neurosciences Dynamics in Complex Brain Networks

CERN Document Server

Graben, Peter beim; Thiel, Marco; Kurths, Jürgen

2008-01-01

Computational Neuroscience is a burgeoning field of research where only the combined effort of neuroscientists, biologists, psychologists, physicists, mathematicians, computer scientists, engineers and other specialists, e.g. from linguistics and medicine, seem to be able to expand the limits of our knowledge. The present volume is an introduction, largely from the physicists' perspective, to the subject matter with in-depth contributions by system neuroscientists. A conceptual model for complex networks of neurons is introduced that incorporates many important features of the real brain, such as various types of neurons, various brain areas, inhibitory and excitatory coupling and the plasticity of the network. The computational implementation on supercomputers, which is introduced and discussed in detail in this book, will enable the readers to modify and adapt the algortihm for their own research. Worked-out examples of applications are presented for networks of Morris-Lecar neurons to model the cortical co...

Symbolic simulation of engineering systems on a supercomputer

International Nuclear Information System (INIS)

Ragheb, M.; Gvillo, D.; Makowitz, H.

1986-01-01

Model-Based Production-Rule systems for analysis are developed for the symbolic simulation of Complex Engineering systems on a CRAY X-MP Supercomputer. The Fault-Tree and Event-Tree Analysis methodologies from Systems-Analysis are used for problem representation and are coupled to the Rule-Based System Paradigm from Knowledge Engineering to provide modelling of engineering devices. Modelling is based on knowledge of the structure and function of the device rather than on human expertise alone. To implement the methodology, we developed a production-Rule Analysis System that uses both backward-chaining and forward-chaining: HAL-1986. The inference engine uses an Induction-Deduction-Oriented antecedent-consequent logic and is programmed in Portable Standard Lisp (PSL). The inference engine is general and can accommodate general modifications and additions to the knowledge base. The methodologies used will be demonstrated using a model for the identification of faults, and subsequent recovery from abnormal situations in Nuclear Reactor Safety Analysis. The use of the exposed methodologies for the prognostication of future device responses under operational and accident conditions using coupled symbolic and procedural programming is discussed

Micro-mechanical Simulations of Soils using Massively Parallel Supercomputers

Directory of Open Access Journals (Sweden)

David W. Washington

2004-06-01

Full Text Available In this research a computer program, Trubal version 1.51, based on the Discrete Element Method was converted to run on a Connection Machine (CM-5,a massively parallel supercomputer with 512 nodes, to expedite the computational times of simulating Geotechnical boundary value problems. The dynamic memory algorithm in Trubal program did not perform efficiently in CM-2 machine with the Single Instruction Multiple Data (SIMD architecture. This was due to the communication overhead involving global array reductions, global array broadcast and random data movement. Therefore, a dynamic memory algorithm in Trubal program was converted to a static memory arrangement and Trubal program was successfully converted to run on CM-5 machines. The converted program was called "TRUBAL for Parallel Machines (TPM." Simulating two physical triaxial experiments and comparing simulation results with Trubal simulations validated the TPM program. With a 512 nodes CM-5 machine TPM produced a nine-fold speedup demonstrating the inherent parallelism within algorithms based on the Discrete Element Method.

Large scale simulations of lattice QCD thermodynamics on Columbia Parallel Supercomputers

International Nuclear Information System (INIS)

Ohta, Shigemi

1989-01-01

The Columbia Parallel Supercomputer project aims at the construction of a parallel processing, multi-gigaflop computer optimized for numerical simulations of lattice QCD. The project has three stages; 16-node, 1/4GF machine completed in April 1985, 64-node, 1GF machine completed in August 1987, and 256-node, 16GF machine now under construction. The machines all share a common architecture; a two dimensional torus formed from a rectangular array of N 1 x N 2 independent and identical processors. A processor is capable of operating in a multi-instruction multi-data mode, except for periods of synchronous interprocessor communication with its four nearest neighbors. Here the thermodynamics simulations on the two working machines are reported. (orig./HSI)

Unique Methodologies for Nano/Micro Manufacturing Job Training Via Desktop Supercomputer Modeling and Simulation

Energy Technology Data Exchange (ETDEWEB)

Kimball, Clyde [Northern Illinois Univ., DeKalb, IL (United States); Karonis, Nicholas [Northern Illinois Univ., DeKalb, IL (United States); Lurio, Laurence [Northern Illinois Univ., DeKalb, IL (United States); Piot, Philippe [Northern Illinois Univ., DeKalb, IL (United States); Xiao, Zhili [Northern Illinois Univ., DeKalb, IL (United States); Glatz, Andreas [Northern Illinois Univ., DeKalb, IL (United States); Pohlman, Nicholas [Northern Illinois Univ., DeKalb, IL (United States); Hou, Minmei [Northern Illinois Univ., DeKalb, IL (United States); Demir, Veysel [Northern Illinois Univ., DeKalb, IL (United States); Song, Jie [Northern Illinois Univ., DeKalb, IL (United States); Duffin, Kirk [Northern Illinois Univ., DeKalb, IL (United States); Johns, Mitrick [Northern Illinois Univ., DeKalb, IL (United States); Sims, Thomas [Northern Illinois Univ., DeKalb, IL (United States); Yin, Yanbin [Northern Illinois Univ., DeKalb, IL (United States)

2012-11-21

This project establishes an initiative in high speed (Teraflop)/large-memory desktop supercomputing for modeling and simulation of dynamic processes important for energy and industrial applications. It provides a training ground for employment of current students in an emerging field with skills necessary to access the large supercomputing systems now present at DOE laboratories. It also provides a foundation for NIU faculty to quantum leap beyond their current small cluster facilities. The funding extends faculty and student capability to a new level of analytic skills with concomitant publication avenues. The components of the Hewlett Packard computer obtained by the DOE funds create a hybrid combination of a Graphics Processing System (12 GPU/Teraflops) and a Beowulf CPU system (144 CPU), the first expandable via the NIU GAEA system to ~60 Teraflops integrated with a 720 CPU Beowulf system. The software is based on access to the NVIDIA/CUDA library and the ability through MATLAB multiple licenses to create additional local programs. A number of existing programs are being transferred to the CPU Beowulf Cluster. Since the expertise necessary to create the parallel processing applications has recently been obtained at NIU, this effort for software development is in an early stage. The educational program has been initiated via formal tutorials and classroom curricula designed for the coming year. Specifically, the cost focus was on hardware acquisitions and appointment of graduate students for a wide range of applications in engineering, physics and computer science.

High Temporal Resolution Mapping of Seismic Noise Sources Using Heterogeneous Supercomputers

Science.gov (United States)

Paitz, P.; Gokhberg, A.; Ermert, L. A.; Fichtner, A.

2017-12-01

The time- and space-dependent distribution of seismic noise sources is becoming a key ingredient of modern real-time monitoring of various geo-systems like earthquake fault zones, volcanoes, geothermal and hydrocarbon reservoirs. We present results of an ongoing research project conducted in collaboration with the Swiss National Supercomputing Centre (CSCS). The project aims at building a service providing seismic noise source maps for Central Europe with high temporal resolution. We use source imaging methods based on the cross-correlation of seismic noise records from all seismic stations available in the region of interest. The service is hosted on the CSCS computing infrastructure; all computationally intensive processing is performed on the massively parallel heterogeneous supercomputer "Piz Daint". The solution architecture is based on the Application-as-a-Service concept to provide the interested researchers worldwide with regular access to the noise source maps. The solution architecture includes the following sub-systems: (1) data acquisition responsible for collecting, on a periodic basis, raw seismic records from the European seismic networks, (2) high-performance noise source mapping application responsible for the generation of source maps using cross-correlation of seismic records, (3) back-end infrastructure for the coordination of various tasks and computations, (4) front-end Web interface providing the service to the end-users and (5) data repository. The noise source mapping itself rests on the measurement of logarithmic amplitude ratios in suitably pre-processed noise correlations, and the use of simplified sensitivity kernels. During the implementation we addressed various challenges, in particular, selection of data sources and transfer protocols, automation and monitoring of daily data downloads, ensuring the required data processing performance, design of a general service-oriented architecture for coordination of various sub-systems, and

De Novo Ultrascale Atomistic Simulations On High-End Parallel Supercomputers

Energy Technology Data Exchange (ETDEWEB)

Nakano, A; Kalia, R K; Nomura, K; Sharma, A; Vashishta, P; Shimojo, F; van Duin, A; Goddard, III, W A; Biswas, R; Srivastava, D; Yang, L H

2006-09-04

We present a de novo hierarchical simulation framework for first-principles based predictive simulations of materials and their validation on high-end parallel supercomputers and geographically distributed clusters. In this framework, high-end chemically reactive and non-reactive molecular dynamics (MD) simulations explore a wide solution space to discover microscopic mechanisms that govern macroscopic material properties, into which highly accurate quantum mechanical (QM) simulations are embedded to validate the discovered mechanisms and quantify the uncertainty of the solution. The framework includes an embedded divide-and-conquer (EDC) algorithmic framework for the design of linear-scaling simulation algorithms with minimal bandwidth complexity and tight error control. The EDC framework also enables adaptive hierarchical simulation with automated model transitioning assisted by graph-based event tracking. A tunable hierarchical cellular decomposition parallelization framework then maps the O(N) EDC algorithms onto Petaflops computers, while achieving performance tunability through a hierarchy of parameterized cell data/computation structures, as well as its implementation using hybrid Grid remote procedure call + message passing + threads programming. High-end computing platforms such as IBM BlueGene/L, SGI Altix 3000 and the NSF TeraGrid provide an excellent test grounds for the framework. On these platforms, we have achieved unprecedented scales of quantum-mechanically accurate and well validated, chemically reactive atomistic simulations--1.06 billion-atom fast reactive force-field MD and 11.8 million-atom (1.04 trillion grid points) quantum-mechanical MD in the framework of the EDC density functional theory on adaptive multigrids--in addition to 134 billion-atom non-reactive space-time multiresolution MD, with the parallel efficiency as high as 0.998 on 65,536 dual-processor BlueGene/L nodes. We have also achieved an automated execution of hierarchical QM

An Optimized Parallel FDTD Topology for Challenging Electromagnetic Simulations on Supercomputers

Directory of Open Access Journals (Sweden)

Shugang Jiang

2015-01-01

Full Text Available It may not be a challenge to run a Finite-Difference Time-Domain (FDTD code for electromagnetic simulations on a supercomputer with more than 10 thousands of CPU cores; however, to make FDTD code work with the highest efficiency is a challenge. In this paper, the performance of parallel FDTD is optimized through MPI (message passing interface virtual topology, based on which a communication model is established. The general rules of optimal topology are presented according to the model. The performance of the method is tested and analyzed on three high performance computing platforms with different architectures in China. Simulations including an airplane with a 700-wavelength wingspan, and a complex microstrip antenna array with nearly 2000 elements are performed very efficiently using a maximum of 10240 CPU cores.

Development of a high performance eigensolver on the peta-scale next generation supercomputer system

International Nuclear Information System (INIS)

Imamura, Toshiyuki; Yamada, Susumu; Machida, Masahiko

2010-01-01

For the present supercomputer systems, a multicore and multisocket processors are necessary to build a system, and choice of interconnection is essential. In addition, for effective development of a new code, high performance, scalable, and reliable numerical software is one of the key items. ScaLAPACK and PETSc are well-known software on distributed memory parallel computer systems. It is needless to say that highly tuned software towards new architecture like many-core processors must be chosen for real computation. In this study, we present a high-performance and high-scalable eigenvalue solver towards the next-generation supercomputer system, so called 'K-computer' system. We have developed two versions, the standard version (eigen s) and enhanced performance version (eigen sx), which are developed on the T2K cluster system housed at University of Tokyo. Eigen s employs the conventional algorithms; Householder tridiagonalization, divide and conquer (DC) algorithm, and Householder back-transformation. They are carefully implemented with blocking technique and flexible two-dimensional data-distribution to reduce the overhead of memory traffic and data transfer, respectively. Eigen s performs excellently on the T2K system with 4096 cores (theoretical peak is 37.6 TFLOPS), and it shows fine performance 3.0 TFLOPS with a two hundred thousand dimensional matrix. The enhanced version, eigen sx, uses more advanced algorithms; the narrow-band reduction algorithm, DC for band matrices, and the block Householder back-transformation with WY-representation. Even though this version is still on a test stage, it shows 4.7 TFLOPS with the same dimensional matrix on eigen s. (author)

Watson will see you now: a supercomputer to help clinicians make informed treatment decisions.

Science.gov (United States)

Doyle-Lindrud, Susan

2015-02-01

IBM has collaborated with several cancer care providers to develop and train the IBM supercomputer Watson to help clinicians make informed treatment decisions. When a patient is seen in clinic, the oncologist can input all of the clinical information into the computer system. Watson will then review all of the data and recommend treatment options based on the latest evidence and guidelines. Once the oncologist makes the treatment decision, this information can be sent directly to the insurance company for approval. Watson has the ability to standardize care and accelerate the approval process, a benefit to the healthcare provider and the patient.

Affordable and accurate large-scale hybrid-functional calculations on GPU-accelerated supercomputers

Science.gov (United States)

Ratcliff, Laura E.; Degomme, A.; Flores-Livas, José A.; Goedecker, Stefan; Genovese, Luigi

2018-03-01

Performing high accuracy hybrid functional calculations for condensed matter systems containing a large number of atoms is at present computationally very demanding or even out of reach if high quality basis sets are used. We present a highly optimized multiple graphics processing unit implementation of the exact exchange operator which allows one to perform fast hybrid functional density-functional theory (DFT) calculations with systematic basis sets without additional approximations for up to a thousand atoms. With this method hybrid DFT calculations of high quality become accessible on state-of-the-art supercomputers within a time-to-solution that is of the same order of magnitude as traditional semilocal-GGA functionals. The method is implemented in a portable open-source library.

Adventures in supercomputing: An innovative program for high school teachers

Energy Technology Data Exchange (ETDEWEB)

Oliver, C.E.; Hicks, H.R.; Summers, B.G. [Oak Ridge National Lab., TN (United States); Staten, D.G. [Wartburg Central High School, TN (United States)

1994-12-31

Within the realm of education, seldom does an innovative program become available with the potential to change an educator`s teaching methodology. Adventures in Supercomputing (AiS), sponsored by the U.S. Department of Energy (DOE), is such a program. It is a program for high school teachers that changes the teacher paradigm from a teacher-directed approach of teaching to a student-centered approach. {open_quotes}A student-centered classroom offers better opportunities for development of internal motivation, planning skills, goal setting and perseverance than does the traditional teacher-directed mode{close_quotes}. Not only is the process of teaching changed, but the cross-curricula integration within the AiS materials is remarkable. Written from a teacher`s perspective, this paper will describe the AiS program and its effects on teachers and students, primarily at Wartburg Central High School, in Wartburg, Tennessee. The AiS program in Tennessee is sponsored by Oak Ridge National Laboratory (ORNL).

Re-inventing electromagnetics - Supercomputing solution of Maxwell's equations via direct time integration on space grids

International Nuclear Information System (INIS)

Taflove, A.

1992-01-01

This paper summarizes the present state and future directions of applying finite-difference and finite-volume time-domain techniques for Maxwell's equations on supercomputers to model complex electromagnetic wave interactions with structures. Applications so far have been dominated by radar cross section technology, but by no means are limited to this area. In fact, the gains we have made place us on the threshold of being able to make tremendous contributions to non-defense electronics and optical technology. Some of the most interesting research in these commercial areas is summarized. 47 refs

Earth and environmental science in the 1980's: Part 1: Environmental data systems, supercomputer facilities and networks

Science.gov (United States)

1986-01-01

Overview descriptions of on-line environmental data systems, supercomputer facilities, and networks are presented. Each description addresses the concepts of content, capability, and user access relevant to the point of view of potential utilization by the Earth and environmental science community. The information on similar systems or facilities is presented in parallel fashion to encourage and facilitate intercomparison. In addition, summary sheets are given for each description, and a summary table precedes each section.

Efficient development of memory bounded geo-applications to scale on modern supercomputers

Science.gov (United States)

Räss, Ludovic; Omlin, Samuel; Licul, Aleksandar; Podladchikov, Yuri; Herman, Frédéric

2016-04-01

Numerical modeling is an actual key tool in the area of geosciences. The current challenge is to solve problems that are multi-physics and for which the length scale and the place of occurrence might not be known in advance. Also, the spatial extend of the investigated domain might strongly vary in size, ranging from millimeters for reactive transport to kilometers for glacier erosion dynamics. An efficient way to proceed is to develop simple but robust algorithms that perform well and scale on modern supercomputers and permit therefore very high-resolution simulations. We propose an efficient approach to solve memory bounded real-world applications on modern supercomputers architectures. We optimize the software to run on our newly acquired state-of-the-art GPU cluster "octopus". Our approach shows promising preliminary results on important geodynamical and geomechanical problematics: we have developed a Stokes solver for glacier flow and a poromechanical solver including complex rheologies for nonlinear waves in stressed rocks porous rocks. We solve the system of partial differential equations on a regular Cartesian grid and use an iterative finite difference scheme with preconditioning of the residuals. The MPI communication happens only locally (point-to-point); this method is known to scale linearly by construction. The "octopus" GPU cluster, which we use for the computations, has been designed to achieve maximal data transfer throughput at minimal hardware cost. It is composed of twenty compute nodes, each hosting four Nvidia Titan X GPU accelerators. These high-density nodes are interconnected with a parallel (dual-rail) FDR InfiniBand network. Our efforts show promising preliminary results for the different physics investigated. The glacier flow solver achieves good accuracy in the relevant benchmarks and the coupled poromechanical solver permits to explain previously unresolvable focused fluid flow as a natural outcome of the porosity setup. In both cases

Research center Juelich to install Germany's most powerful supercomputer new IBM System for science and research will achieve 5.8 trillion computations per second

CERN Multimedia

2002-01-01

"The Research Center Juelich, Germany, and IBM today announced that they have signed a contract for the delivery and installation of a new IBM supercomputer at the Central Institute for Applied Mathematics" (1/2 page).

MEGADOCK 4.0: an ultra-high-performance protein-protein docking software for heterogeneous supercomputers.

Science.gov (United States)

Ohue, Masahito; Shimoda, Takehiro; Suzuki, Shuji; Matsuzaki, Yuri; Ishida, Takashi; Akiyama, Yutaka

2014-11-15

The application of protein-protein docking in large-scale interactome analysis is a major challenge in structural bioinformatics and requires huge computing resources. In this work, we present MEGADOCK 4.0, an FFT-based docking software that makes extensive use of recent heterogeneous supercomputers and shows powerful, scalable performance of >97% strong scaling. MEGADOCK 4.0 is written in C++ with OpenMPI and NVIDIA CUDA 5.0 (or later) and is freely available to all academic and non-profit users at: http://www.bi.cs.titech.ac.jp/megadock. akiyama@cs.titech.ac.jp Supplementary data are available at Bioinformatics online. © The Author 2014. Published by Oxford University Press.

Wavelet transform-vector quantization compression of supercomputer ocean model simulation output

Energy Technology Data Exchange (ETDEWEB)

Bradley, J N; Brislawn, C M

1992-11-12

We describe a new procedure for efficient compression of digital information for storage and transmission purposes. The algorithm involves a discrete wavelet transform subband decomposition of the data set, followed by vector quantization of the wavelet transform coefficients using application-specific vector quantizers. The new vector quantizer design procedure optimizes the assignment of both memory resources and vector dimensions to the transform subbands by minimizing an exponential rate-distortion functional subject to constraints on both overall bit-rate and encoder complexity. The wavelet-vector quantization method, which originates in digital image compression. is applicable to the compression of other multidimensional data sets possessing some degree of smoothness. In this paper we discuss the use of this technique for compressing the output of supercomputer simulations of global climate models. The data presented here comes from Semtner-Chervin global ocean models run at the National Center for Atmospheric Research and at the Los Alamos Advanced Computing Laboratory.

Computational fluid dynamics: complex flows requiring supercomputers. January 1975-July 1988 (Citations from the INSPEC: Information Services for the Physics and Engineering Communities data base). Report for January 1975-July 1988

International Nuclear Information System (INIS)

1988-08-01

This bibliography contains citations concerning computational fluid dynamics (CFD), a new method in computational science to perform complex flow simulations in three dimensions. Applications include aerodynamic design and analysis for aircraft, rockets, and missiles, and automobiles; heat-transfer studies; and combustion processes. Included are references to supercomputers, array processors, and parallel processors where needed for complete, integrated design. Also included are software packages and grid-generation techniques required to apply CFD numerical solutions. Numerical methods for fluid dynamics, not requiring supercomputers, are found in a separate published search. (Contains 83 citations fully indexed and including a title list.)

Performance Characteristics of Hybrid MPI/OpenMP Scientific Applications on a Large-Scale Multithreaded BlueGene/Q Supercomputer

KAUST Repository

Wu, Xingfu; Taylor, Valerie

2013-01-01

In this paper, we investigate the performance characteristics of five hybrid MPI/OpenMP scientific applications (two NAS Parallel benchmarks Multi-Zone SP-MZ and BT-MZ, an earthquake simulation PEQdyna, an aerospace application PMLB and a 3D particle-in-cell application GTC) on a large-scale multithreaded Blue Gene/Q supercomputer at Argonne National laboratory, and quantify the performance gap resulting from using different number of threads per node. We use performance tools and MPI profile and trace libraries available on the supercomputer to analyze and compare the performance of these hybrid scientific applications with increasing the number OpenMP threads per node, and find that increasing the number of threads to some extent saturates or worsens performance of these hybrid applications. For the strong-scaling hybrid scientific applications such as SP-MZ, BT-MZ, PEQdyna and PLMB, using 32 threads per node results in much better application efficiency than using 64 threads per node, and as increasing the number of threads per node, the FPU (Floating Point Unit) percentage decreases, and the MPI percentage (except PMLB) and IPC (Instructions per cycle) per core (except BT-MZ) increase. For the weak-scaling hybrid scientific application such as GTC, the performance trend (relative speedup) is very similar with increasing number of threads per node no matter how many nodes (32, 128, 512) are used. © 2013 IEEE.

Performance Characteristics of Hybrid MPI/OpenMP Scientific Applications on a Large-Scale Multithreaded BlueGene/Q Supercomputer

KAUST Repository

Wu, Xingfu

2013-07-01

In this paper, we investigate the performance characteristics of five hybrid MPI/OpenMP scientific applications (two NAS Parallel benchmarks Multi-Zone SP-MZ and BT-MZ, an earthquake simulation PEQdyna, an aerospace application PMLB and a 3D particle-in-cell application GTC) on a large-scale multithreaded Blue Gene/Q supercomputer at Argonne National laboratory, and quantify the performance gap resulting from using different number of threads per node. We use performance tools and MPI profile and trace libraries available on the supercomputer to analyze and compare the performance of these hybrid scientific applications with increasing the number OpenMP threads per node, and find that increasing the number of threads to some extent saturates or worsens performance of these hybrid applications. For the strong-scaling hybrid scientific applications such as SP-MZ, BT-MZ, PEQdyna and PLMB, using 32 threads per node results in much better application efficiency than using 64 threads per node, and as increasing the number of threads per node, the FPU (Floating Point Unit) percentage decreases, and the MPI percentage (except PMLB) and IPC (Instructions per cycle) per core (except BT-MZ) increase. For the weak-scaling hybrid scientific application such as GTC, the performance trend (relative speedup) is very similar with increasing number of threads per node no matter how many nodes (32, 128, 512) are used. © 2013 IEEE.

Research to application: Supercomputing trends for the 90's - Opportunities for interdisciplinary computations

International Nuclear Information System (INIS)

Shankar, V.

1991-01-01

The progression of supercomputing is reviewed from the point of view of computational fluid dynamics (CFD), and multidisciplinary problems impacting the design of advanced aerospace configurations are addressed. The application of full potential and Euler equations to transonic and supersonic problems in the 70s and early 80s is outlined, along with Navier-Stokes computations widespread during the late 80s and early 90s. Multidisciplinary computations currently in progress are discussed, including CFD and aeroelastic coupling for both static and dynamic flexible computations, CFD, aeroelastic, and controls coupling for flutter suppression and active control, and the development of a computational electromagnetics technology based on CFD methods. Attention is given to computational challenges standing in a way of the concept of establishing a computational environment including many technologies. 40 refs

Car2x with software defined networks, network functions virtualization and supercomputers technical and scientific preparations for the Amsterdam Arena telecoms fieldlab

NARCIS (Netherlands)

Meijer R.J.; Cushing R.; De Laat C.; Jackson P.; Klous S.; Koning R.; Makkes M.X.; Meerwijk A.

2015-01-01

In the invited talk 'Car2x with SDN, NFV and supercomputers' we report about how our past work with SDN [1, 2] allows the design of a smart mobility fieldlab in the huge parking lot the Amsterdam Arena. We explain how we can engineer and test software that handle the complex conditions of the Car2X

A user-friendly web portal for T-Coffee on supercomputers

Directory of Open Access Journals (Sweden)

Koetsier Jos

2011-05-01

Full Text Available Abstract Background Parallel T-Coffee (PTC was the first parallel implementation of the T-Coffee multiple sequence alignment tool. It is based on MPI and RMA mechanisms. Its purpose is to reduce the execution time of the large-scale sequence alignments. It can be run on distributed memory clusters allowing users to align data sets consisting of hundreds of proteins within a reasonable time. However, most of the potential users of this tool are not familiar with the use of grids or supercomputers. Results In this paper we show how PTC can be easily deployed and controlled on a super computer architecture using a web portal developed using Rapid. Rapid is a tool for efficiently generating standardized portlets for a wide range of applications and the approach described here is generic enough to be applied to other applications, or to deploy PTC on different HPC environments. Conclusions The PTC portal allows users to upload a large number of sequences to be aligned by the parallel version of TC that cannot be aligned by a single machine due to memory and execution time constraints. The web portal provides a user-friendly solution.

Portable implementation model for CFD simulations. Application to hybrid CPU/GPU supercomputers

Science.gov (United States)

Oyarzun, Guillermo; Borrell, Ricard; Gorobets, Andrey; Oliva, Assensi

2017-10-01

Nowadays, high performance computing (HPC) systems experience a disruptive moment with a variety of novel architectures and frameworks, without any clarity of which one is going to prevail. In this context, the portability of codes across different architectures is of major importance. This paper presents a portable implementation model based on an algebraic operational approach for direct numerical simulation (DNS) and large eddy simulation (LES) of incompressible turbulent flows using unstructured hybrid meshes. The strategy proposed consists in representing the whole time-integration algorithm using only three basic algebraic operations: sparse matrix-vector product, a linear combination of vectors and dot product. The main idea is based on decomposing the nonlinear operators into a concatenation of two SpMV operations. This provides high modularity and portability. An exhaustive analysis of the proposed implementation for hybrid CPU/GPU supercomputers has been conducted with tests using up to 128 GPUs. The main objective consists in understanding the challenges of implementing CFD codes on new architectures.

EDF's experience with supercomputing and challenges ahead - towards multi-physics and multi-scale approaches

International Nuclear Information System (INIS)

Delbecq, J.M.; Banner, D.

2003-01-01

Nuclear power plants are a major asset of the EDF company. To remain so, in particular in a context of deregulation, competitiveness, safety and public acceptance are three conditions. These stakes apply both to existing plants and to future reactors. The purpose of the presentation is to explain how supercomputing can help EDF to satisfy these requirements. Three examples are described in detail: ensuring optimal use of nuclear fuel under wholly safe conditions, understanding and simulating the material deterioration mechanisms and moving forward with numerical simulation for the performance of EDF's activities. In conclusion, a broader vision of EDF long term R and D in the field of numerical simulation is given and especially of five challenges taken up by EDF together with its industrial and scientific partners. (author)

SOFTWARE FOR SUPERCOMPUTER SKIF “ProLit-lC” and “ProNRS-lC” FOR FOUNDRY AND METALLURGICAL PRODUCTIONS

Directory of Open Access Journals (Sweden)

A. N. Chichko

2008-01-01

Full Text Available The data of modeling on supercomputer system SKIF of technological process of  molds filling by means of computer system 'ProLIT-lc', and also data of modeling of the steel pouring process by means ofTroNRS-lc'are presented. The influence of number of  processors of  multinuclear computer system SKIF on acceleration and time of  modeling of technological processes, connected with production of castings and slugs, is shown.

Palacios and Kitten : high performance operating systems for scalable virtualized and native supercomputing.

Energy Technology Data Exchange (ETDEWEB)

Widener, Patrick (University of New Mexico); Jaconette, Steven (Northwestern University); Bridges, Patrick G. (University of New Mexico); Xia, Lei (Northwestern University); Dinda, Peter (Northwestern University); Cui, Zheng.; Lange, John (Northwestern University); Hudson, Trammell B.; Levenhagen, Michael J.; Pedretti, Kevin Thomas Tauke; Brightwell, Ronald Brian

2009-09-01

Palacios and Kitten are new open source tools that enable applications, whether ported or not, to achieve scalable high performance on large machines. They provide a thin layer over the hardware to support both full-featured virtualized environments and native code bases. Kitten is an OS under development at Sandia that implements a lightweight kernel architecture to provide predictable behavior and increased flexibility on large machines, while also providing Linux binary compatibility. Palacios is a VMM that is under development at Northwestern University and the University of New Mexico. Palacios, which can be embedded into Kitten and other OSes, supports existing, unmodified applications and operating systems by using virtualization that leverages hardware technologies. We describe the design and implementation of both Kitten and Palacios. Our benchmarks show that they provide near native, scalable performance. Palacios and Kitten provide an incremental path to using supercomputer resources that is not performance-compromised.

Performance Analysis and Scaling Behavior of the Terrestrial Systems Modeling Platform TerrSysMP in Large-Scale Supercomputing Environments

Science.gov (United States)

Kollet, S. J.; Goergen, K.; Gasper, F.; Shresta, P.; Sulis, M.; Rihani, J.; Simmer, C.; Vereecken, H.

2013-12-01

In studies of the terrestrial hydrologic, energy and biogeochemical cycles, integrated multi-physics simulation platforms take a central role in characterizing non-linear interactions, variances and uncertainties of system states and fluxes in reciprocity with observations. Recently developed integrated simulation platforms attempt to honor the complexity of the terrestrial system across multiple time and space scales from the deeper subsurface including groundwater dynamics into the atmosphere. Technically, this requires the coupling of atmospheric, land surface, and subsurface-surface flow models in supercomputing environments, while ensuring a high-degree of efficiency in the utilization of e.g., standard Linux clusters and massively parallel resources. A systematic performance analysis including profiling and tracing in such an application is crucial in the understanding of the runtime behavior, to identify optimum model settings, and is an efficient way to distinguish potential parallel deficiencies. On sophisticated leadership-class supercomputers, such as the 28-rack 5.9 petaFLOP IBM Blue Gene/Q 'JUQUEEN' of the Jülich Supercomputing Centre (JSC), this is a challenging task, but even more so important, when complex coupled component models are to be analysed. Here we want to present our experience from coupling, application tuning (e.g. 5-times speedup through compiler optimizations), parallel scaling and performance monitoring of the parallel Terrestrial Systems Modeling Platform TerrSysMP. The modeling platform consists of the weather prediction system COSMO of the German Weather Service; the Community Land Model, CLM of NCAR; and the variably saturated surface-subsurface flow code ParFlow. The model system relies on the Multiple Program Multiple Data (MPMD) execution model where the external Ocean-Atmosphere-Sea-Ice-Soil coupler (OASIS3) links the component models. TerrSysMP has been instrumented with the performance analysis tool Scalasca and analyzed

Assessment techniques for a learning-centered curriculum: evaluation design for adventures in supercomputing

Energy Technology Data Exchange (ETDEWEB)

Helland, B. [Ames Lab., IA (United States); Summers, B.G. [Oak Ridge National Lab., TN (United States)

1996-09-01

As the classroom paradigm shifts from being teacher-centered to being learner-centered, student assessments are evolving from typical paper and pencil testing to other methods of evaluation. Students should be probed for understanding, reasoning, and critical thinking abilities rather than their ability to return memorized facts. The assessment of the Department of Energy`s pilot program, Adventures in Supercomputing (AiS), offers one example of assessment techniques developed for learner-centered curricula. This assessment has employed a variety of methods to collect student data. Methods of assessment used were traditional testing, performance testing, interviews, short questionnaires via email, and student presentations of projects. The data obtained from these sources have been analyzed by a professional assessment team at the Center for Children and Technology. The results have been used to improve the AiS curriculum and establish the quality of the overall AiS program. This paper will discuss the various methods of assessment used and the results.

The BlueGene/L Supercomputer and Quantum ChromoDynamics

International Nuclear Information System (INIS)

Vranas, P; Soltz, R

2006-01-01

In summary our update contains: (1) Perfect speedup sustaining 19.3% of peak for the Wilson D D-slash Dirac operator. (2) Measurements of the full Conjugate Gradient (CG) inverter that inverts the Dirac operator. The CG inverter contains two global sums over the entire machine. Nevertheless, our measurements retain perfect speedup scaling demonstrating the robustness of our methods. (3) We ran on the largest BG/L system, the LLNL 64 rack BG/L supercomputer, and obtained a sustained speed of 59.1 TFlops. Furthermore, the speedup scaling of the Dirac operator and of the CG inverter are perfect all the way up to the full size of the machine, 131,072 cores (please see Figure II). The local lattice is rather small (4 x 4 x 4 x 16) while the total lattice has been a lattice QCD vision for thermodynamic studies (a total of 128 x 128 x 256 x 32 lattice sites). This speed is about five times larger compared to the speed we quoted in our submission. As we have pointed out in our paper QCD is notoriously sensitive to network and memory latencies, has a relatively high communication to computation ratio which can not be overlapped in BGL in virtual node mode, and as an application is in a class of its own. The above results are thrilling to us and a 30 year long dream for lattice QCD

Modeling radiative transport in ICF plasmas on an IBM SP2 supercomputer

International Nuclear Information System (INIS)

Johansen, J.A.; MacFarlane, J.J.; Moses, G.A.

1995-01-01

At the University of Wisconsin-Madison the authors have integrated a collisional-radiative-equilibrium model into their CONRAD radiation-hydrodynamics code. This integrated package allows them to accurately simulate the transport processes involved in ICF plasmas; including the important effects of self-absorption of line-radiation. However, as they increase the amount of atomic structure utilized in their transport models, the computational demands increase nonlinearly. In an attempt to meet this increased computational demand, they have recently embarked on a mission to parallelize the CONRAD program. The parallel CONRAD development is being performed on an IBM SP2 supercomputer. The parallelism is based on a message passing paradigm, and is being implemented using PVM. At the present time they have determined that approximately 70% of the sequential program can be executed in parallel. Accordingly, they expect that the parallel version will yield a speedup on the order of three times that of the sequential version. This translates into only 10 hours of execution time for the parallel version, whereas the sequential version required 30 hours

Solving sparse linear least squares problems on some supercomputers by using large dense blocks

DEFF Research Database (Denmark)

Hansen, Per Christian; Ostromsky, T; Sameh, A

1997-01-01

technique is preferable to sparse matrix technique when the matrices are not large, because the high computational speed compensates fully the disadvantages of using more arithmetic operations and more storage. For very large matrices the computations must be organized as a sequence of tasks in each......Efficient subroutines for dense matrix computations have recently been developed and are available on many high-speed computers. On some computers the speed of many dense matrix operations is near to the peak-performance. For sparse matrices storage and operations can be saved by operating only...... and storing only nonzero elements. However, the price is a great degradation of the speed of computations on supercomputers (due to the use of indirect addresses, to the need to insert new nonzeros in the sparse storage scheme, to the lack of data locality, etc.). On many high-speed computers a dense matrix...

Accelerating Science Impact through Big Data Workflow Management and Supercomputing

Directory of Open Access Journals (Sweden)

De K.

2016-01-01

Full Text Available The Large Hadron Collider (LHC, operating at the international CERN Laboratory in Geneva, Switzerland, is leading Big Data driven scientific explorations. ATLAS, one of the largest collaborations ever assembled in the the history of science, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. To manage the workflow for all data processing on hundreds of data centers the PanDA (Production and Distributed AnalysisWorkload Management System is used. An ambitious program to expand PanDA to all available computing resources, including opportunistic use of commercial and academic clouds and Leadership Computing Facilities (LCF, is realizing within BigPanDA and megaPanDA projects. These projects are now exploring how PanDA might be used for managing computing jobs that run on supercomputers including OLCF’s Titan and NRC-KI HPC2. The main idea is to reuse, as much as possible, existing components of the PanDA system that are already deployed on the LHC Grid for analysis of physics data. The next generation of PanDA will allow many data-intensive sciences employing a variety of computing platforms to benefit from ATLAS experience and proven tools in highly scalable processing.

HEP Computing Tools, Grid and Supercomputers for Genome Sequencing Studies

Science.gov (United States)

De, K.; Klimentov, A.; Maeno, T.; Mashinistov, R.; Novikov, A.; Poyda, A.; Tertychnyy, I.; Wenaus, T.

2017-10-01

PanDA - Production and Distributed Analysis Workload Management System has been developed to address ATLAS experiment at LHC data processing and analysis challenges. Recently PanDA has been extended to run HEP scientific applications on Leadership Class Facilities and supercomputers. The success of the projects to use PanDA beyond HEP and Grid has drawn attention from other compute intensive sciences such as bioinformatics. Recent advances of Next Generation Genome Sequencing (NGS) technology led to increasing streams of sequencing data that need to be processed, analysed and made available for bioinformaticians worldwide. Analysis of genomes sequencing data using popular software pipeline PALEOMIX can take a month even running it on the powerful computer resource. In this paper we will describe the adaptation the PALEOMIX pipeline to run it on a distributed computing environment powered by PanDA. To run pipeline we split input files into chunks which are run separately on different nodes as separate inputs for PALEOMIX and finally merge output file, it is very similar to what it done by ATLAS to process and to simulate data. We dramatically decreased the total walltime because of jobs (re)submission automation and brokering within PanDA. Using software tools developed initially for HEP and Grid can reduce payload execution time for Mammoths DNA samples from weeks to days.

EDF's experience with supercomputing and challenges ahead - towards multi-physics and multi-scale approaches

Energy Technology Data Exchange (ETDEWEB)

Delbecq, J.M.; Banner, D. [Electricite de France (EDF)- R and D Division, 92 - Clamart (France)

2003-07-01

Nuclear power plants are a major asset of the EDF company. To remain so, in particular in a context of deregulation, competitiveness, safety and public acceptance are three conditions. These stakes apply both to existing plants and to future reactors. The purpose of the presentation is to explain how supercomputing can help EDF to satisfy these requirements. Three examples are described in detail: ensuring optimal use of nuclear fuel under wholly safe conditions, understanding and simulating the material deterioration mechanisms and moving forward with numerical simulation for the performance of EDF's activities. In conclusion, a broader vision of EDF long term R and D in the field of numerical simulation is given and especially of five challenges taken up by EDF together with its industrial and scientific partners. (author)

Performance Evaluation of an Intel Haswell- and Ivy Bridge-Based Supercomputer Using Scientific and Engineering Applications

Science.gov (United States)

Saini, Subhash; Hood, Robert T.; Chang, Johnny; Baron, John

2016-01-01

We present a performance evaluation conducted on a production supercomputer of the Intel Xeon Processor E5- 2680v3, a twelve-core implementation of the fourth-generation Haswell architecture, and compare it with Intel Xeon Processor E5-2680v2, an Ivy Bridge implementation of the third-generation Sandy Bridge architecture. Several new architectural features have been incorporated in Haswell including improvements in all levels of the memory hierarchy as well as improvements to vector instructions and power management. We critically evaluate these new features of Haswell and compare with Ivy Bridge using several low-level benchmarks including subset of HPCC, HPCG and four full-scale scientific and engineering applications. We also present a model to predict the performance of HPCG and Cart3D within 5%, and Overflow within 10% accuracy.

369 TFlop/s molecular dynamics simulations on the Roadrunner general-purpose heterogeneous supercomputer

Energy Technology Data Exchange (ETDEWEB)

Swaminarayan, Sriram [Los Alamos National Laboratory; Germann, Timothy C [Los Alamos National Laboratory; Kadau, Kai [Los Alamos National Laboratory; Fossum, Gordon C [IBM CORPORATION

2008-01-01

The authors present timing and performance numbers for a short-range parallel molecular dynamics (MD) code, SPaSM, that has been rewritten for the heterogeneous Roadrunner supercomputer. Each Roadrunner compute node consists of two AMD Opteron dual-core microprocessors and four PowerXCell 8i enhanced Cell microprocessors, so that there are four MPI ranks per node, each with one Opteron and one Cell. The interatomic forces are computed on the Cells (each with one PPU and eight SPU cores), while the Opterons are used to direct inter-rank communication and perform I/O-heavy periodic analysis, visualization, and checkpointing tasks. The performance measured for our initial implementation of a standard Lennard-Jones pair potential benchmark reached a peak of 369 Tflop/s double-precision floating-point performance on the full Roadrunner system (27.7% of peak), corresponding to 124 MFlop/Watt/s at a price of approximately 3.69 MFlops/dollar. They demonstrate an initial target application, the jetting and ejection of material from a shocked surface.

New Mexico High School Supercomputing Challenge, 1990--1995: Five years of making a difference to students, teachers, schools, and communities. Progress report

Energy Technology Data Exchange (ETDEWEB)

Foster, M.; Kratzer, D.

1996-02-01

The New Mexico High School Supercomputing Challenge is an academic program dedicated to increasing interest in science and math among high school students by introducing them to high performance computing. This report provides a summary and evaluation of the first five years of the program, describes the program and shows the impact that it has had on high school students, their teachers, and their communities. Goals and objectives are reviewed and evaluated, growth and development of the program are analyzed, and future directions are discussed.

Parallel simulation of tsunami inundation on a large-scale supercomputer

Science.gov (United States)

Oishi, Y.; Imamura, F.; Sugawara, D.

2013-12-01

An accurate prediction of tsunami inundation is important for disaster mitigation purposes. One approach is to approximate the tsunami wave source through an instant inversion analysis using real-time observation data (e.g., Tsushima et al., 2009) and then use the resulting wave source data in an instant tsunami inundation simulation. However, a bottleneck of this approach is the large computational cost of the non-linear inundation simulation and the computational power of recent massively parallel supercomputers is helpful to enable faster than real-time execution of a tsunami inundation simulation. Parallel computers have become approximately 1000 times faster in 10 years (www.top500.org), and so it is expected that very fast parallel computers will be more and more prevalent in the near future. Therefore, it is important to investigate how to efficiently conduct a tsunami simulation on parallel computers. In this study, we are targeting very fast tsunami inundation simulations on the K computer, currently the fastest Japanese supercomputer, which has a theoretical peak performance of 11.2 PFLOPS. One computing node of the K computer consists of 1 CPU with 8 cores that share memory, and the nodes are connected through a high-performance torus-mesh network. The K computer is designed for distributed-memory parallel computation, so we have developed a parallel tsunami model. Our model is based on TUNAMI-N2 model of Tohoku University, which is based on a leap-frog finite difference method. A grid nesting scheme is employed to apply high-resolution grids only at the coastal regions. To balance the computation load of each CPU in the parallelization, CPUs are first allocated to each nested layer in proportion to the number of grid points of the nested layer. Using CPUs allocated to each layer, 1-D domain decomposition is performed on each layer. In the parallel computation, three types of communication are necessary: (1) communication to adjacent neighbours for the

Reliability Lessons Learned From GPU Experience With The Titan Supercomputer at Oak Ridge Leadership Computing Facility

Energy Technology Data Exchange (ETDEWEB)

Gallarno, George [Christian Brothers University; Rogers, James H [ORNL; Maxwell, Don E [ORNL

2015-01-01

The high computational capability of graphics processing units (GPUs) is enabling and driving the scientific discovery process at large-scale. The world s second fastest supercomputer for open science, Titan, has more than 18,000 GPUs that computational scientists use to perform scientific simu- lations and data analysis. Understanding of GPU reliability characteristics, however, is still in its nascent stage since GPUs have only recently been deployed at large-scale. This paper presents a detailed study of GPU errors and their impact on system operations and applications, describing experiences with the 18,688 GPUs on the Titan supercom- puter as well as lessons learned in the process of efficient operation of GPUs at scale. These experiences are helpful to HPC sites which already have large-scale GPU clusters or plan to deploy GPUs in the future.

A criticality safety analysis code using a vectorized Monte Carlo method on the HITAC S-810 supercomputer

International Nuclear Information System (INIS)

Morimoto, Y.; Maruyama, H.

1987-01-01

A vectorized Monte Carlo criticality safety analysis code has been developed on the vector supercomputer HITAC S-810. In this code, a multi-particle tracking algorithm was adopted for effective utilization of the vector processor. A flight analysis with pseudo-scattering was developed to reduce the computational time needed for flight analysis, which represents the bulk of computational time. This new algorithm realized a speed-up of factor 1.5 over the conventional flight analysis. The code also adopted the multigroup cross section constants library of the Bodarenko type with 190 groups, with 132 groups being for fast and epithermal regions and 58 groups being for the thermal region. Evaluation work showed that this code reproduce the experimental results to an accuracy of about 1 % for the effective neutron multiplication factor. (author)

Benchmarking Further Single Board Computers for Building a Mini Supercomputer for Simulation of Telecommunication Systems

Directory of Open Access Journals (Sweden)

Gábor Lencse

2016-01-01

Full Text Available Parallel Discrete Event Simulation (PDES with the conservative synchronization method can be efficiently used for the performance analysis of telecommunication systems because of their good lookahead properties. For PDES, a cost effective execution platform may be built by using single board computers (SBCs, which offer relatively high computation capacity compared to their price or power consumption and especially to the space they take up. A benchmarking method is proposed and its operation is demonstrated by benchmarking ten different SBCs, namely Banana Pi, Beaglebone Black, Cubieboard2, Odroid-C1+, Odroid-U3+, Odroid-XU3 Lite, Orange Pi Plus, Radxa Rock Lite, Raspberry Pi Model B+, and Raspberry Pi 2 Model B+. Their benchmarking results are compared to find out which one should be used for building a mini supercomputer for parallel discrete-event simulation of telecommunication systems. The SBCs are also used to build a heterogeneous cluster and the performance of the cluster is tested, too.

Fast and Accurate Simulation of the Cray XMT Multithreaded Supercomputer

Energy Technology Data Exchange (ETDEWEB)

Villa, Oreste; Tumeo, Antonino; Secchi, Simone; Manzano Franco, Joseph B.

2012-12-31

Irregular applications, such as data mining and analysis or graph-based computations, show unpredictable memory/network access patterns and control structures. Highly multithreaded architectures with large processor counts, like the Cray MTA-1, MTA-2 and XMT, appear to address their requirements better than commodity clusters. However, the research on highly multithreaded systems is currently limited by the lack of adequate architectural simulation infrastructures due to issues such as size of the machines, memory footprint, simulation speed, accuracy and customization. At the same time, Shared-memory MultiProcessors (SMPs) with multi-core processors have become an attractive platform to simulate large scale machines. In this paper, we introduce a cycle-level simulator of the highly multithreaded Cray XMT supercomputer. The simulator runs unmodified XMT applications. We discuss how we tackled the challenges posed by its development, detailing the techniques introduced to make the simulation as fast as possible while maintaining a high accuracy. By mapping XMT processors (ThreadStorm with 128 hardware threads) to host computing cores, the simulation speed remains constant as the number of simulated processors increases, up to the number of available host cores. The simulator supports zero-overhead switching among different accuracy levels at run-time and includes a network model that takes into account contention. On a modern 48-core SMP host, our infrastructure simulates a large set of irregular applications 500 to 2000 times slower than real time when compared to a 128-processor XMT, while remaining within 10\\% of accuracy. Emulation is only from 25 to 200 times slower than real time.

PFLOTRAN: Reactive Flow & Transport Code for Use on Laptops to Leadership-Class Supercomputers

Energy Technology Data Exchange (ETDEWEB)

Hammond, Glenn E.; Lichtner, Peter C.; Lu, Chuan; Mills, Richard T.

2012-04-18

PFLOTRAN, a next-generation reactive flow and transport code for modeling subsurface processes, has been designed from the ground up to run efficiently on machines ranging from leadership-class supercomputers to laptops. Based on an object-oriented design, the code is easily extensible to incorporate additional processes. It can interface seamlessly with Fortran 9X, C and C++ codes. Domain decomposition parallelism is employed, with the PETSc parallel framework used to manage parallel solvers, data structures and communication. Features of the code include a modular input file, implementation of high-performance I/O using parallel HDF5, ability to perform multiple realization simulations with multiple processors per realization in a seamless manner, and multiple modes for multiphase flow and multicomponent geochemical transport. Chemical reactions currently implemented in the code include homogeneous aqueous complexing reactions and heterogeneous mineral precipitation/dissolution, ion exchange, surface complexation and a multirate kinetic sorption model. PFLOTRAN has demonstrated petascale performance using 2{sup 17} processor cores with over 2 billion degrees of freedom. Accomplishments achieved to date include applications to the Hanford 300 Area and modeling CO{sub 2} sequestration in deep geologic formations.

NASA's Climate in a Box: Desktop Supercomputing for Open Scientific Model Development

Science.gov (United States)

Wojcik, G. S.; Seablom, M. S.; Lee, T. J.; McConaughy, G. R.; Syed, R.; Oloso, A.; Kemp, E. M.; Greenseid, J.; Smith, R.

2009-12-01

NASA's High Performance Computing Portfolio in cooperation with its Modeling, Analysis, and Prediction program intends to make its climate and earth science models more accessible to a larger community. A key goal of this effort is to open the model development and validation process to the scientific community at large such that a natural selection process is enabled and results in a more efficient scientific process. One obstacle to others using NASA models is the complexity of the models and the difficulty in learning how to use them. This situation applies not only to scientists who regularly use these models but also non-typical users who may want to use the models such as scientists from different domains, policy makers, and teachers. Another obstacle to the use of these models is that access to high performance computing (HPC) accounts, from which the models are implemented, can be restrictive with long wait times in job queues and delays caused by an arduous process of obtaining an account, especially for foreign nationals. This project explores the utility of using desktop supercomputers in providing a complete ready-to-use toolkit of climate research products to investigators and on demand access to an HPC system. One objective of this work is to pre-package NASA and NOAA models so that new users will not have to spend significant time porting the models. In addition, the prepackaged toolkit will include tools, such as workflow, visualization, social networking web sites, and analysis tools, to assist users in running the models and analyzing the data. The system architecture to be developed will allow for automatic code updates for each user and an effective means with which to deal with data that are generated. We plan to investigate several desktop systems, but our work to date has focused on a Cray CX1. Currently, we are investigating the potential capabilities of several non-traditional development environments. While most NASA and NOAA models are

Evaluating the networking characteristics of the Cray XC-40 Intel Knights Landing-based Cori supercomputer at NERSC

Energy Technology Data Exchange (ETDEWEB)

Doerfler, Douglas [Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Austin, Brian [Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Cook, Brandon [Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Deslippe, Jack [Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States); Kandalla, Krishna [Cray Inc, Bloomington, MN (United States); Mendygral, Peter [Cray Inc, Bloomington, MN (United States)

2017-09-12

There are many potential issues associated with deploying the Intel Xeon Phi™ (code named Knights Landing [KNL]) manycore processor in a large-scale supercomputer. One in particular is the ability to fully utilize the high-speed communications network, given that the serial performance of a Xeon Phi TM core is a fraction of a Xeon®core. In this paper, we take a look at the trade-offs associated with allocating enough cores to fully utilize the Aries high-speed network versus cores dedicated to computation, e.g., the trade-off between MPI and OpenMP. In addition, we evaluate new features of Cray MPI in support of KNL, such as internode optimizations. We also evaluate one-sided programming models such as Unified Parallel C. We quantify the impact of the above trade-offs and features using a suite of National Energy Research Scientific Computing Center applications.

Using the LANSCE irradiation facility to predict the number of fatal soft errors in one of the world's fastest supercomputers

International Nuclear Information System (INIS)

Michalak, S.E.; Harris, K.W.; Hengartner, N.W.; Takala, B.E.; Wender, S.A.

2005-01-01

Los Alamos National Laboratory (LANL) is home to the Los Alamos Neutron Science Center (LANSCE). LANSCE is a unique facility because its neutron spectrum closely mimics the neutron spectrum at terrestrial and aircraft altitudes, but is many times more intense. Thus, LANSCE provides an ideal setting for accelerated testing of semiconductor and other devices that are susceptible to cosmic ray induced neutrons. Many industrial companies use LANSCE to estimate device susceptibility to cosmic ray induced neutrons, and it has also been used to test parts from one of LANL's supercomputers, the ASC (Advanced Simulation and Computing Program) Q. This paper discusses our use of the LANSCE facility to study components in Q including a comparison with failure data from Q

Parallel supercomputing: Advanced methods, algorithms, and software for large-scale linear and nonlinear problems

Energy Technology Data Exchange (ETDEWEB)

Carey, G.F.; Young, D.M.

1993-12-31

The program outlined here is directed to research on methods, algorithms, and software for distributed parallel supercomputers. Of particular interest are finite element methods and finite difference methods together with sparse iterative solution schemes for scientific and engineering computations of very large-scale systems. Both linear and nonlinear problems will be investigated. In the nonlinear case, applications with bifurcation to multiple solutions will be considered using continuation strategies. The parallelizable numerical methods of particular interest are a family of partitioning schemes embracing domain decomposition, element-by-element strategies, and multi-level techniques. The methods will be further developed incorporating parallel iterative solution algorithms with associated preconditioners in parallel computer software. The schemes will be implemented on distributed memory parallel architectures such as the CRAY MPP, Intel Paragon, the NCUBE3, and the Connection Machine. We will also consider other new architectures such as the Kendall-Square (KSQ) and proposed machines such as the TERA. The applications will focus on large-scale three-dimensional nonlinear flow and reservoir problems with strong convective transport contributions. These are legitimate grand challenge class computational fluid dynamics (CFD) problems of significant practical interest to DOE. The methods developed and algorithms will, however, be of wider interest.

Visualization at supercomputing centers: the tale of little big iron and the three skinny guys.

Science.gov (United States)

Bethel, E W; van Rosendale, J; Southard, D; Gaither, K; Childs, H; Brugger, E; Ahern, S

2011-01-01

Supercomputing centers are unique resources that aim to enable scientific knowledge discovery by employing large computational resources-the "Big Iron." Design, acquisition, installation, and management of the Big Iron are carefully planned and monitored. Because these Big Iron systems produce a tsunami of data, it's natural to colocate the visualization and analysis infrastructure. This infrastructure consists of hardware (Little Iron) and staff (Skinny Guys). Our collective experience suggests that design, acquisition, installation, and management of the Little Iron and Skinny Guys doesn't receive the same level of treatment as that of the Big Iron. This article explores the following questions about the Little Iron: How should we size the Little Iron to adequately support visualization and analysis of data coming off the Big Iron? What sort of capabilities must it have? Related questions concern the size of visualization support staff: How big should a visualization program be-that is, how many Skinny Guys should it have? What should the staff do? How much of the visualization should be provided as a support service, and how much should applications scientists be expected to do on their own?

The design and implementation of cost-effective algorithms for direct solution of banded linear systems on the vector processor system 32 supercomputer

Science.gov (United States)

Samba, A. S.

1985-01-01

The problem of solving banded linear systems by direct (non-iterative) techniques on the Vector Processor System (VPS) 32 supercomputer is considered. Two efficient direct methods for solving banded linear systems on the VPS 32 are described. The vector cyclic reduction (VCR) algorithm is discussed in detail. The performance of the VCR on a three parameter model problem is also illustrated. The VCR is an adaptation of the conventional point cyclic reduction algorithm. The second direct method is the Customized Reduction of Augmented Triangles' (CRAT). CRAT has the dominant characteristics of an efficient VPS 32 algorithm. CRAT is tailored to the pipeline architecture of the VPS 32 and as a consequence the algorithm is implicitly vectorizable.

Sandia`s network for Supercomputing `94: Linking the Los Alamos, Lawrence Livermore, and Sandia National Laboratories using switched multimegabit data service

Energy Technology Data Exchange (ETDEWEB)

Vahle, M.O.; Gossage, S.A.; Brenkosh, J.P. [Sandia National Labs., Albuquerque, NM (United States). Advanced Networking Integration Dept.

1995-01-01

Supercomputing `94, a high-performance computing and communications conference, was held November 14th through 18th, 1994 in Washington DC. For the past four years, Sandia National Laboratories has used this conference to showcase and focus its communications and networking endeavors. At the 1994 conference, Sandia built a Switched Multimegabit Data Service (SMDS) network running at 44.736 megabits per second linking its private SMDS network between its facilities in Albuquerque, New Mexico and Livermore, California to the convention center in Washington, D.C. For the show, the network was also extended from Sandia, New Mexico to Los Alamos National Laboratory and from Sandia, California to Lawrence Livermore National Laboratory. This paper documents and describes this network and how it was used at the conference.

Combining density functional theory calculations, supercomputing, and data-driven methods to design new materials (Conference Presentation)

Science.gov (United States)

Jain, Anubhav

2017-04-01

Density functional theory (DFT) simulations solve for the electronic structure of materials starting from the Schrödinger equation. Many case studies have now demonstrated that researchers can often use DFT to design new compounds in the computer (e.g., for batteries, catalysts, and hydrogen storage) before synthesis and characterization in the lab. In this talk, I will focus on how DFT calculations can be executed on large supercomputing resources in order to generate very large data sets on new materials for functional applications. First, I will briefly describe the Materials Project, an effort at LBNL that has virtually characterized over 60,000 materials using DFT and has shared the results with over 17,000 registered users. Next, I will talk about how such data can help discover new materials, describing how preliminary computational screening led to the identification and confirmation of a new family of bulk AMX2 thermoelectric compounds with measured zT reaching 0.8. I will outline future plans for how such data-driven methods can be used to better understand the factors that control thermoelectric behavior, e.g., for the rational design of electronic band structures, in ways that are different from conventional approaches.

A Parallel Supercomputer Implementation of a Biological Inspired Neural Network and its use for Pattern Recognition

International Nuclear Information System (INIS)

De Ladurantaye, Vincent; Lavoie, Jean; Bergeron, Jocelyn; Parenteau, Maxime; Lu Huizhong; Pichevar, Ramin; Rouat, Jean

2012-01-01

A parallel implementation of a large spiking neural network is proposed and evaluated. The neural network implements the binding by synchrony process using the Oscillatory Dynamic Link Matcher (ODLM). Scalability, speed and performance are compared for 2 implementations: Message Passing Interface (MPI) and Compute Unified Device Architecture (CUDA) running on clusters of multicore supercomputers and NVIDIA graphical processing units respectively. A global spiking list that represents at each instant the state of the neural network is described. This list indexes each neuron that fires during the current simulation time so that the influence of their spikes are simultaneously processed on all computing units. Our implementation shows a good scalability for very large networks. A complex and large spiking neural network has been implemented in parallel with success, thus paving the road towards real-life applications based on networks of spiking neurons. MPI offers a better scalability than CUDA, while the CUDA implementation on a GeForce GTX 285 gives the best cost to performance ratio. When running the neural network on the GTX 285, the processing speed is comparable to the MPI implementation on RQCHP's Mammouth parallel with 64 notes (128 cores).

High Performance Simulation of Large-Scale Red Sea Ocean Bottom Seismic Data on the Supercomputer Shaheen II

KAUST Repository

Tonellot, Thierry

2017-02-27

A combination of both shallow and deepwater, plus islands and coral reefs, are some of the main features contributing to the complexity of subsalt seismic exploration in the Red Sea transition zone. These features often result in degrading effects on seismic images. State-of-the-art ocean bottom acquisition technologies are therefore required to record seismic data with optimal fold and offset, as well as advanced processing and imaging techniques. Numerical simulations of such complex seismic data can help improve acquisition design and also help in customizing, validating and benchmarking the processing and imaging workflows that will be applied on the field data. Subsequently, realistic simulation of wave propagation is a computationally intensive process requiring a realistic model and an efficient 3D wave equation solver. Large-scale computing resources are also required to meet turnaround time compatible with a production time frame. In this work, we present the numerical simulation of an ocean bottom seismic survey to be acquired in the Red Sea transition zone starting in summer 2016. The survey\\'s acquisition geometry comprises nearly 300,000 unique shot locations and 21,000 unique receiver locations, covering about 760 km2. Using well log measurements and legacy 2D seismic lines in this area, a 3D P-wave velocity model was built, with a maximum depth of 7 km. The model was sampled at 10 m in each direction, resulting in more than 5 billion cells. Wave propagation in this model was performed using a 3D finite difference solver in the time domain based on a staggered grid velocity-pressure formulation of acoustodynamics. To ensure that the resulting data could be generated sufficiently fast, the King Abdullah University of Science and Technology (KAUST) supercomputer Shaheen II Cray XC40 was used. A total of 21,000 three-component (pressure and vertical and horizontal velocity) common receiver gathers with a 50 Hz maximum frequency were computed in less

High Performance Simulation of Large-Scale Red Sea Ocean Bottom Seismic Data on the Supercomputer Shaheen II

KAUST Repository

Tonellot, Thierry; Etienne, Vincent; Gashawbeza, Ewenet; Curiel, Emesto Sandoval; Khan, Azizur; Feki, Saber; Kortas, Samuel

2017-01-01

A combination of both shallow and deepwater, plus islands and coral reefs, are some of the main features contributing to the complexity of subsalt seismic exploration in the Red Sea transition zone. These features often result in degrading effects on seismic images. State-of-the-art ocean bottom acquisition technologies are therefore required to record seismic data with optimal fold and offset, as well as advanced processing and imaging techniques. Numerical simulations of such complex seismic data can help improve acquisition design and also help in customizing, validating and benchmarking the processing and imaging workflows that will be applied on the field data. Subsequently, realistic simulation of wave propagation is a computationally intensive process requiring a realistic model and an efficient 3D wave equation solver. Large-scale computing resources are also required to meet turnaround time compatible with a production time frame. In this work, we present the numerical simulation of an ocean bottom seismic survey to be acquired in the Red Sea transition zone starting in summer 2016. The survey's acquisition geometry comprises nearly 300,000 unique shot locations and 21,000 unique receiver locations, covering about 760 km2. Using well log measurements and legacy 2D seismic lines in this area, a 3D P-wave velocity model was built, with a maximum depth of 7 km. The model was sampled at 10 m in each direction, resulting in more than 5 billion cells. Wave propagation in this model was performed using a 3D finite difference solver in the time domain based on a staggered grid velocity-pressure formulation of acoustodynamics. To ensure that the resulting data could be generated sufficiently fast, the King Abdullah University of Science and Technology (KAUST) supercomputer Shaheen II Cray XC40 was used. A total of 21,000 three-component (pressure and vertical and horizontal velocity) common receiver gathers with a 50 Hz maximum frequency were computed in less than

Use of QUADRICS supercomputer as embedded simulator in emergency management systems; Utilizzo del calcolatore QUADRICS come simulatore in linea in un sistema di gestione delle emergenze

Energy Technology Data Exchange (ETDEWEB)

Bove, R.; Di Costanzo, G.; Ziparo, A. [ENEA, Centro Ricerche Casaccia, Rome (Italy). Dip. Energia

1996-07-01

The experience related to the implementation of a MRBT, atmospheric spreading model with a short duration releasing, are reported. This model was implemented on a QUADRICS-Q1 supercomputer. First is reported a description of the MRBT model. It is an analytical model to study the speadings of light gases realised in the atmosphere cause incidental releasing. The solution of diffusion equation is Gaussian like. It yield the concentration of pollutant substance released. The concentration is function of space and time. Thus the QUADRICS architecture is introduced. And the implementation of the model is described. At the end it will be consider the integration of the QUADRICS-based model as simulator in a emergency management system.

Summaries of research and development activities by using supercomputer system of JAEA in FY2015. April 1, 2015 - March 31, 2016

International Nuclear Information System (INIS)

2017-01-01

Japan Atomic Energy Agency (JAEA) conducts research and development (R and D) in various fields related to nuclear power as a comprehensive institution of nuclear energy R and Ds, and utilizes computational science and technology in many activities. As shown in the fact that about 20 percent of papers published by JAEA are concerned with R and D using computational science, the supercomputer system of JAEA has become an important infrastructure to support computational science and technology. In FY2015, the system was used for R and D aiming to restore Fukushima (nuclear plant decommissioning and environmental restoration) as a priority issue, as well as for JAEA's major projects such as Fast Reactor Cycle System, Fusion R and D and Quantum Beam Science. This report presents a great number of R and D results accomplished by using the system in FY2015, as well as user support, operational records and overviews of the system, and so on. (author)

Summaries of research and development activities by using supercomputer system of JAEA in FY2014. April 1, 2014 - March 31, 2015

International Nuclear Information System (INIS)

2016-02-01

Japan Atomic Energy Agency (JAEA) conducts research and development (R and D) in various fields related to nuclear power as a comprehensive institution of nuclear energy R and Ds, and utilizes computational science and technology in many activities. As shown in the fact that about 20 percent of papers published by JAEA are concerned with R and D using computational science, the supercomputer system of JAEA has become an important infrastructure to support computational science and technology. In FY2014, the system was used for R and D aiming to restore Fukushima (nuclear plant decommissioning and environmental restoration) as a priority issue, as well as for JAEA's major projects such as Fast Reactor Cycle System, Fusion R and D and Quantum Beam Science. This report presents a great number of R and D results accomplished by using the system in FY2014, as well as user support, operational records and overviews of the system, and so on. (author)

Summaries of research and development activities by using supercomputer system of JAEA in FY2013. April 1, 2013 - March 31, 2014

International Nuclear Information System (INIS)

2015-02-01

Japan Atomic Energy Agency (JAEA) conducts research and development (R and D) in various fields related to nuclear power as a comprehensive institution of nuclear energy R and Ds, and utilizes computational science and technology in many activities. About 20 percent of papers published by JAEA are concerned with R and D using computational science, the supercomputer system of JAEA has become an important infrastructure to support computational science and technology utilization. In FY2013, the system was used not only for JAEA's major projects such as Fast Reactor Cycle System, Fusion R and D and Quantum Beam Science, but also for R and D aiming to restore Fukushima (nuclear plant decommissioning and environmental restoration) as a priority issue. This report presents a great amount of R and D results accomplished by using the system in FY2013, as well as user support, operational records and overviews of the system, and so on. (author)

Summaries of research and development activities by using supercomputer system of JAEA in FY2012. April 1, 2012 - March 31, 2013

International Nuclear Information System (INIS)

2014-01-01

Japan Atomic Energy Agency (JAEA) conducts research and development (R and D) in various fields related to nuclear power as a comprehensive institution of nuclear energy R and Ds, and utilizes computational science and technology in many activities. As more than 20 percent of papers published by JAEA are concerned with R and D using computational science, the supercomputer system of JAEA has become an important infrastructure to support computational science and technology utilization. In FY2012, the system was used not only for JAEA's major projects such as Fast Reactor Cycle System, Fusion R and D and Quantum Beam Science, but also for R and D aiming to restore Fukushima (nuclear plant decommissioning and environmental restoration) as apriority issue. This report presents a great amount of R and D results accomplished by using the system in FY2012, as well as user support, operational records and overviews of the system, and so on. (author)

Summaries of research and development activities by using supercomputer system of JAEA in FY2011. April 1, 2011 - March 31, 2012

International Nuclear Information System (INIS)

2013-01-01

Japan Atomic Energy Agency (JAEA) conducts research and development (R and D) in various fields related to nuclear power as a comprehensive institution of nuclear energy R and Ds, and utilizes computational science and technology in many activities. As more than 20 percent of papers published by JAEA are concerned with R and D using computational science, the supercomputer system of JAEA has become an important infrastructure to support computational science and technology utilization. In FY2011, the system was used for analyses of the accident at the Fukushima Daiichi Nuclear Power Station and establishment of radioactive decontamination plan, as well as the JAEA's major projects such as Fast Reactor Cycle System, Fusion R and D and Quantum Beam Science. This report presents a great amount of R and D results accomplished by using the system in FY2011, as well as user support structure, operational records and overviews of the system, and so on. (author)

Getting To Exascale: Applying Novel Parallel Programming Models To Lab Applications For The Next Generation Of Supercomputers

Energy Technology Data Exchange (ETDEWEB)

Dube, Evi [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States); Shereda, Charles [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States); Nau, Lee [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States); Harris, Lance [Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)

2010-09-27

As supercomputing moves toward exascale, node architectures will change significantly. CPU core counts on nodes will increase by an order of magnitude or more. Heterogeneous architectures will become more commonplace, with GPUs or FPGAs providing additional computational power. Novel programming models may make better use of on-node parallelism in these new architectures than do current models. In this paper we examine several of these novel models – UPC, CUDA, and OpenCL –to determine their suitability to LLNL scientific application codes. Our study consisted of several phases: We conducted interviews with code teams and selected two codes to port; We learned how to program in the new models and ported the codes; We debugged and tuned the ported applications; We measured results, and documented our findings. We conclude that UPC is a challenge for porting code, Berkeley UPC is not very robust, and UPC is not suitable as a general alternative to OpenMP for a number of reasons. CUDA is well supported and robust but is a proprietary NVIDIA standard, while OpenCL is an open standard. Both are well suited to a specific set of application problems that can be run on GPUs, but some problems are not suited to GPUs. Further study of the landscape of novel models is recommended.

Argonne Leadership Computing Facility 2011 annual report : Shaping future supercomputing.

Energy Technology Data Exchange (ETDEWEB)

Papka, M.; Messina, P.; Coffey, R.; Drugan, C. (LCF)

2012-08-16

The ALCF's Early Science Program aims to prepare key applications for the architecture and scale of Mira and to solidify libraries and infrastructure that will pave the way for other future production applications. Two billion core-hours have been allocated to 16 Early Science projects on Mira. The projects, in addition to promising delivery of exciting new science, are all based on state-of-the-art, petascale, parallel applications. The project teams, in collaboration with ALCF staff and IBM, have undertaken intensive efforts to adapt their software to take advantage of Mira's Blue Gene/Q architecture, which, in a number of ways, is a precursor to future high-performance-computing architecture. The Argonne Leadership Computing Facility (ALCF) enables transformative science that solves some of the most difficult challenges in biology, chemistry, energy, climate, materials, physics, and other scientific realms. Users partnering with ALCF staff have reached research milestones previously unattainable, due to the ALCF's world-class supercomputing resources and expertise in computation science. In 2011, the ALCF's commitment to providing outstanding science and leadership-class resources was honored with several prestigious awards. Research on multiscale brain blood flow simulations was named a Gordon Bell Prize finalist. Intrepid, the ALCF's BG/P system, ranked No. 1 on the Graph 500 list for the second consecutive year. The next-generation BG/Q prototype again topped the Green500 list. Skilled experts at the ALCF enable researchers to conduct breakthrough science on the Blue Gene system in key ways. The Catalyst Team matches project PIs with experienced computational scientists to maximize and accelerate research in their specific scientific domains. The Performance Engineering Team facilitates the effective use of applications on the Blue Gene system by assessing and improving the algorithms used by applications and the techniques used to

The ASCI Network for SC '99: A Step on the Path to a 100 Gigabit Per Second Supercomputing Network

Energy Technology Data Exchange (ETDEWEB)

PRATT,THOMAS J.; TARMAN,THOMAS D.; MARTINEZ,LUIS M.; MILLER,MARC M.; ADAMS,ROGER L.; CHEN,HELEN Y.; BRANDT,JAMES M.; WYCKOFF,PETER S.

2000-07-24

This document highlights the Discom{sup 2}'s Distance computing and communication team activities at the 1999 Supercomputing conference in Portland, Oregon. This conference is sponsored by the IEEE and ACM. Sandia, Lawrence Livermore and Los Alamos National laboratories have participated in this conference for eleven years. For the last four years the three laboratories have come together at the conference under the DOE's ASCI, Accelerated Strategic Computing Initiatives rubric. Communication support for the ASCI exhibit is provided by the ASCI DISCOM{sup 2} project. The DISCOM{sup 2} communication team uses this forum to demonstrate and focus communication and networking developments within the community. At SC 99, DISCOM built a prototype of the next generation ASCI network demonstrated remote clustering techniques, demonstrated the capabilities of the emerging Terabit Routers products, demonstrated the latest technologies for delivering visualization data to the scientific users, and demonstrated the latest in encryption methods including IP VPN technologies and ATM encryption research. The authors also coordinated the other production networking activities within the booth and between their demonstration partners on the exhibit floor. This paper documents those accomplishments, discusses the details of their implementation, and describes how these demonstrations support Sandia's overall strategies in ASCI networking.

New generation of docking programs: Supercomputer validation of force fields and quantum-chemical methods for docking.

Science.gov (United States)

Sulimov, Alexey V; Kutov, Danil C; Katkova, Ekaterina V; Ilin, Ivan S; Sulimov, Vladimir B

2017-11-01

Discovery of new inhibitors of the protein associated with a given disease is the initial and most important stage of the whole process of the rational development of new pharmaceutical substances. New inhibitors block the active site of the target protein and the disease is cured. Computer-aided molecular modeling can considerably increase effectiveness of new inhibitors development. Reliable predictions of the target protein inhibition by a small molecule, ligand, is defined by the accuracy of docking programs. Such programs position a ligand in the target protein and estimate the protein-ligand binding energy. Positioning accuracy of modern docking programs is satisfactory. However, the accuracy of binding energy calculations is too low to predict good inhibitors. For effective application of docking programs to new inhibitors development the accuracy of binding energy calculations should be higher than 1kcal/mol. Reasons of limited accuracy of modern docking programs are discussed. One of the most important aspects limiting this accuracy is imperfection of protein-ligand energy calculations. Results of supercomputer validation of several force fields and quantum-chemical methods for docking are presented. The validation was performed by quasi-docking as follows. First, the low energy minima spectra of 16 protein-ligand complexes were found by exhaustive minima search in the MMFF94 force field. Second, energies of the lowest 8192 minima are recalculated with CHARMM force field and PM6-D3H4X and PM7 quantum-chemical methods for each complex. The analysis of minima energies reveals the docking positioning accuracies of the PM7 and PM6-D3H4X quantum-chemical methods and the CHARMM force field are close to one another and they are better than the positioning accuracy of the MMFF94 force field. Copyright © 2017 Elsevier Inc. All rights reserved.

Scalable geocomputation: evolving an environmental model building platform from single-core to supercomputers

Science.gov (United States)

Schmitz, Oliver; de Jong, Kor; Karssenberg, Derek

2017-04-01

There is an increasing demand to run environmental models on a big scale: simulations over large areas at high resolution. The heterogeneity of available computing hardware such as multi-core CPUs, GPUs or supercomputer potentially provides significant computing power to fulfil this demand. However, this requires detailed knowledge of the underlying hardware, parallel algorithm design and the implementation thereof in an efficient system programming language. Domain scientists such as hydrologists or ecologists often lack this specific software engineering knowledge, their emphasis is (and should be) on exploratory building and analysis of simulation models. As a result, models constructed by domain specialists mostly do not take full advantage of the available hardware. A promising solution is to separate the model building activity from software engineering by offering domain specialists a model building framework with pre-programmed building blocks that they combine to construct a model. The model building framework, consequently, needs to have built-in capabilities to make full usage of the available hardware. Developing such a framework providing understandable code for domain scientists and being runtime efficient at the same time poses several challenges on developers of such a framework. For example, optimisations can be performed on individual operations or the whole model, or tasks need to be generated for a well-balanced execution without explicitly knowing the complexity of the domain problem provided by the modeller. Ideally, a modelling framework supports the optimal use of available hardware whichsoever combination of model building blocks scientists use. We demonstrate our ongoing work on developing parallel algorithms for spatio-temporal modelling and demonstrate 1) PCRaster, an environmental software framework (http://www.pcraster.eu) providing spatio-temporal model building blocks and 2) parallelisation of about 50 of these building blocks using

Grassroots Supercomputing

CERN Multimedia

Buchanan, Mark

2005-01-01

What started out as a way for SETI to plow through its piles or radio-signal data from deep space has turned into a powerful research tool as computer users acrosse the globe donate their screen-saver time to projects as diverse as climate-change prediction, gravitational-wave searches, and protein folding (4 pages)

MaMiCo: Transient multi-instance molecular-continuum flow simulation on supercomputers

Science.gov (United States)

Neumann, Philipp; Bian, Xin

2017-11-01

We present extensions of the macro-micro-coupling tool MaMiCo, which was designed to couple continuum fluid dynamics solvers with discrete particle dynamics. To enable local extraction of smooth flow field quantities especially on rather short time scales, sampling over an ensemble of molecular dynamics simulations is introduced. We provide details on these extensions including the transient coupling algorithm, open boundary forcing, and multi-instance sampling. Furthermore, we validate the coupling in Couette flow using different particle simulation software packages and particle models, i.e. molecular dynamics and dissipative particle dynamics. Finally, we demonstrate the parallel scalability of the molecular-continuum simulations by using up to 65 536 compute cores of the supercomputer Shaheen II located at KAUST. Program Files doi:http://dx.doi.org/10.17632/w7rgdrhb85.1 Licensing provisions: BSD 3-clause Programming language: C, C++ External routines/libraries: For compiling: SCons, MPI (optional) Subprograms used: ESPResSo, LAMMPS, ls1 mardyn, waLBerla For installation procedures of the MaMiCo interfaces, see the README files in the respective code directories located in coupling/interface/impl. Journal reference of previous version: P. Neumann, H. Flohr, R. Arora, P. Jarmatz, N. Tchipev, H.-J. Bungartz. MaMiCo: Software design for parallel molecular-continuum flow simulations, Computer Physics Communications 200: 324-335, 2016 Does the new version supersede the previous version?: Yes. The functionality of the previous version is completely retained in the new version. Nature of problem: Coupled molecular-continuum simulation for multi-resolution fluid dynamics: parts of the domain are resolved by molecular dynamics or another particle-based solver whereas large parts are covered by a mesh-based CFD solver, e.g. a lattice Boltzmann automaton. Solution method: We couple existing MD and CFD solvers via MaMiCo (macro-micro coupling tool). Data exchange and

Three-dimensional kinetic simulations of whistler turbulence in solar wind on parallel supercomputers

Science.gov (United States)

Chang, Ouliang

The objective of this dissertation is to study the physics of whistler turbulence evolution and its role in energy transport and dissipation in the solar wind plasmas through computational and theoretical investigations. This dissertation presents the first fully three-dimensional (3D) particle-in-cell (PIC) simulations of whistler turbulence forward cascade in a homogeneous, collisionless plasma with a uniform background magnetic field B o, and the first 3D PIC simulation of whistler turbulence with both forward and inverse cascades. Such computationally demanding research is made possible through the use of massively parallel, high performance electromagnetic PIC simulations on state-of-the-art supercomputers. Simulations are carried out to study characteristic properties of whistler turbulence under variable solar wind fluctuation amplitude (epsilon e) and electron beta (betae), relative contributions to energy dissipation and electron heating in whistler turbulence from the quasilinear scenario and the intermittency scenario, and whistler turbulence preferential cascading direction and wavevector anisotropy. The 3D simulations of whistler turbulence exhibit a forward cascade of fluctuations into broadband, anisotropic, turbulent spectrum at shorter wavelengths with wavevectors preferentially quasi-perpendicular to B o. The overall electron heating yields T ∥ > T⊥ for all epsilone and betae values, indicating the primary linear wave-particle interaction is Landau damping. But linear wave-particle interactions play a minor role in shaping the wavevector spectrum, whereas nonlinear wave-wave interactions are overall stronger and faster processes, and ultimately determine the wavevector anisotropy. Simulated magnetic energy spectra as function of wavenumber show a spectral break to steeper slopes, which scales as k⊥lambda e ≃ 1 independent of betae values, where lambdae is electron inertial length, qualitatively similar to solar wind observations. Specific

Algorithms for supercomputers

International Nuclear Information System (INIS)

Alder, B.J.

1986-01-01

Better numerical procedures, improved computational power and additional physical insights have contributed significantly to progress in dealing with classical and quantum statistical mechanics problems. Past developments are discussed and future possibilities outlined

Super-computer architecture

CERN Document Server

Hockney, R W

1977-01-01

This paper examines the design of the top-of-the-range, scientific, number-crunching computers. The market for such computers is not as large as that for smaller machines, but on the other hand it is by no means negligible. The present work-horse machines in this category are the CDC 7600 and IBM 360/195, and over fifty of the former machines have been sold. The types of installation that form the market for such machines are not only the major scientific research laboratories in the major countries-such as Los Alamos, CERN, Rutherford laboratory-but also major universities or university networks. It is also true that, as with sports cars, innovations made to satisfy the top of the market today often become the standard for the medium-scale computer of tomorrow. Hence there is considerable interest in examining present developments in this area. (0 refs).

Algorithms for supercomputers

International Nuclear Information System (INIS)

Alder, B.J.

1985-12-01

Better numerical procedures, improved computational power and additional physical insights have contributed significantly to progress in dealing with classical and quantum statistical mechanics problems. Past developments are discussed and future possibilities outlined

Leveraging HPC resources for High Energy Physics

International Nuclear Information System (INIS)

O'Brien, B; Washbrook, A; Walker, R

2014-01-01

High Performance Computing (HPC) supercomputers provide unprecedented computing power for a diverse range of scientific applications. The most powerful supercomputers now deliver petaflop peak performance with the expectation of 'exascale' technologies available in the next five years. More recent HPC facilities use x86-based architectures managed by Linux-based operating systems which could potentially allow unmodified HEP software to be run on supercomputers. There is now a renewed interest from both the LHC experiments and the HPC community to accommodate data analysis and event simulation production on HPC facilities. This study provides an outline of the challenges faced when incorporating HPC resources for HEP software by using the HECToR supercomputer as a demonstrator.

High Performance Computing in Science and Engineering '02 : Transactions of the High Performance Computing Center

CERN Document Server

Jäger, Willi

2003-01-01

This book presents the state-of-the-art in modeling and simulation on supercomputers. Leading German research groups present their results achieved on high-end systems of the High Performance Computing Center Stuttgart (HLRS) for the year 2002. Reports cover all fields of supercomputing simulation ranging from computational fluid dynamics to computer science. Special emphasis is given to industrially relevant applications. Moreover, by presenting results for both vector sytems and micro-processor based systems the book allows to compare performance levels and usability of a variety of supercomputer architectures. It therefore becomes an indispensable guidebook to assess the impact of the Japanese Earth Simulator project on supercomputing in the years to come.

A high level language for a high performance computer

Science.gov (United States)

Perrott, R. H.

1978-01-01

The proposed computational aerodynamic facility will join the ranks of the supercomputers due to its architecture and increased execution speed. At present, the languages used to program these supercomputers have been modifications of programming languages which were designed many years ago for sequential machines. A new programming language should be developed based on the techniques which have proved valuable for sequential programming languages and incorporating the algorithmic techniques required for these supercomputers. The design objectives for such a language are outlined.

Dal CERN, flusso si dati a una media di 600 megabytes al secondo per dieci giorni consecutivi

CERN Multimedia

2005-01-01

The supercomputer Grid took up successfully its first technologic challenge. Egiht supercomputing centers have supported on internet a continuous flow of data from CERN in Geneva and directed them to seven centers in Europe and United States

Study on the climate system and mass transport by a climate model

International Nuclear Information System (INIS)

Numaguti, A.; Sugata, S.; Takahashi, M.; Nakajima, T.; Sumi, A.

1997-01-01

The Center for Global Environmental Research (CGER), an organ of the National Institute for Environmental Studies of the Environment Agency of Japan, was established in October 1990 to contribute broadly to the scientific understanding of global change, and to the elucidation of and solution for our pressing environmental problems. CGER conducts environmental research from interdisciplinary, multiagency, and international perspective, provides research support facilities such as a supercomputer and databases, and offers its own data from long-term monitoring of the global environment. In March 1992, CGER installed a supercomputer system (NEC SX-3, Model 14) to facilitate research on global change. The system is open to environmental researchers worldwide. Proposed research programs are evaluated by the Supercomputer Steering Committee which consists of leading scientists in climate modeling, atmospheric chemistry, oceanic circulation, and computer science. After project approval, authorization for system usage is provided. In 1995 and 1996, several research proposals were designated as priority research and allocated larger shares of computer resources. The CGER supercomputer monograph report Vol. 3 is a report of priority research of CGER's supercomputer. The report covers the description of CCSR-NIES atmospheric general circulation model, lagragian general circulation based on the time-scale of particle motion, and ability of the CCSR-NIES atmospheric general circulation model in the stratosphere. The results obtained from these three studies are described in three chapters. We hope this report provides you with useful information on the global environmental research conducted on our supercomputer

Massively Parallel QCD

International Nuclear Information System (INIS)

Soltz, R; Vranas, P; Blumrich, M; Chen, D; Gara, A; Giampap, M; Heidelberger, P; Salapura, V; Sexton, J; Bhanot, G

2007-01-01

The theory of the strong nuclear force, Quantum Chromodynamics (QCD), can be numerically simulated from first principles on massively-parallel supercomputers using the method of Lattice Gauge Theory. We describe the special programming requirements of lattice QCD (LQCD) as well as the optimal supercomputer hardware architectures that it suggests. We demonstrate these methods on the BlueGene massively-parallel supercomputer and argue that LQCD and the BlueGene architecture are a natural match. This can be traced to the simple fact that LQCD is a regular lattice discretization of space into lattice sites while the BlueGene supercomputer is a discretization of space into compute nodes, and that both are constrained by requirements of locality. This simple relation is both technologically important and theoretically intriguing. The main result of this paper is the speedup of LQCD using up to 131,072 CPUs on the largest BlueGene/L supercomputer. The speedup is perfect with sustained performance of about 20% of peak. This corresponds to a maximum of 70.5 sustained TFlop/s. At these speeds LQCD and BlueGene are poised to produce the next generation of strong interaction physics theoretical results

Performance Analysis of FEM Algorithmson GPU and Many-Core Architectures

KAUST Repository

Khurram, Rooh; Kortas, Samuel

2015-01-01

-only Exascale systems will be unsustainable, thus accelerators such as graphic processing units (GPUs) and many-integrated-core (MIC) will likely be the integral part of the TOP500 (http://www.top500.org/) supercomputers, beyond 2020. The emerging supercomputer

Hierarchical approach to optimization of parallel matrix multiplication on large-scale platforms

KAUST Repository

Hasanov, Khalid; Quintin, Jean-Noë l; Lastovetsky, Alexey

2014-01-01

-scale parallelism in mind. Indeed, while in 1990s a system with few hundred cores was considered a powerful supercomputer, modern top supercomputers have millions of cores. In this paper, we present a hierarchical approach to optimization of message-passing parallel

Collaborating CPU and GPU for large-scale high-order CFD simulations with complex grids on the TianHe-1A supercomputer

Energy Technology Data Exchange (ETDEWEB)

Xu, Chuanfu, E-mail: xuchuanfu@nudt.edu.cn [College of Computer Science, National University of Defense Technology, Changsha 410073 (China); Deng, Xiaogang; Zhang, Lilun [College of Computer Science, National University of Defense Technology, Changsha 410073 (China); Fang, Jianbin [Parallel and Distributed Systems Group, Delft University of Technology, Delft 2628CD (Netherlands); Wang, Guangxue; Jiang, Yi [State Key Laboratory of Aerodynamics, P.O. Box 211, Mianyang 621000 (China); Cao, Wei; Che, Yonggang; Wang, Yongxian; Wang, Zhenghua; Liu, Wei; Cheng, Xinghua [College of Computer Science, National University of Defense Technology, Changsha 410073 (China)

2014-12-01

Programming and optimizing complex, real-world CFD codes on current many-core accelerated HPC systems is very challenging, especially when collaborating CPUs and accelerators to fully tap the potential of heterogeneous systems. In this paper, with a tri-level hybrid and heterogeneous programming model using MPI + OpenMP + CUDA, we port and optimize our high-order multi-block structured CFD software HOSTA on the GPU-accelerated TianHe-1A supercomputer. HOSTA adopts two self-developed high-order compact definite difference schemes WCNS and HDCS that can simulate flows with complex geometries. We present a dual-level parallelization scheme for efficient multi-block computation on GPUs and perform particular kernel optimizations for high-order CFD schemes. The GPU-only approach achieves a speedup of about 1.3 when comparing one Tesla M2050 GPU with two Xeon X5670 CPUs. To achieve a greater speedup, we collaborate CPU and GPU for HOSTA instead of using a naive GPU-only approach. We present a novel scheme to balance the loads between the store-poor GPU and the store-rich CPU. Taking CPU and GPU load balance into account, we improve the maximum simulation problem size per TianHe-1A node for HOSTA by 2.3×, meanwhile the collaborative approach can improve the performance by around 45% compared to the GPU-only approach. Further, to scale HOSTA on TianHe-1A, we propose a gather/scatter optimization to minimize PCI-e data transfer times for ghost and singularity data of 3D grid blocks, and overlap the collaborative computation and communication as far as possible using some advanced CUDA and MPI features. Scalability tests show that HOSTA can achieve a parallel efficiency of above 60% on 1024 TianHe-1A nodes. With our method, we have successfully simulated an EET high-lift airfoil configuration containing 800M cells and China's large civil airplane configuration containing 150M cells. To our best knowledge, those are the largest-scale CPU–GPU collaborative simulations

Collaborating CPU and GPU for large-scale high-order CFD simulations with complex grids on the TianHe-1A supercomputer

International Nuclear Information System (INIS)

Xu, Chuanfu; Deng, Xiaogang; Zhang, Lilun; Fang, Jianbin; Wang, Guangxue; Jiang, Yi; Cao, Wei; Che, Yonggang; Wang, Yongxian; Wang, Zhenghua; Liu, Wei; Cheng, Xinghua

2014-01-01

Programming and optimizing complex, real-world CFD codes on current many-core accelerated HPC systems is very challenging, especially when collaborating CPUs and accelerators to fully tap the potential of heterogeneous systems. In this paper, with a tri-level hybrid and heterogeneous programming model using MPI + OpenMP + CUDA, we port and optimize our high-order multi-block structured CFD software HOSTA on the GPU-accelerated TianHe-1A supercomputer. HOSTA adopts two self-developed high-order compact definite difference schemes WCNS and HDCS that can simulate flows with complex geometries. We present a dual-level parallelization scheme for efficient multi-block computation on GPUs and perform particular kernel optimizations for high-order CFD schemes. The GPU-only approach achieves a speedup of about 1.3 when comparing one Tesla M2050 GPU with two Xeon X5670 CPUs. To achieve a greater speedup, we collaborate CPU and GPU for HOSTA instead of using a naive GPU-only approach. We present a novel scheme to balance the loads between the store-poor GPU and the store-rich CPU. Taking CPU and GPU load balance into account, we improve the maximum simulation problem size per TianHe-1A node for HOSTA by 2.3×, meanwhile the collaborative approach can improve the performance by around 45% compared to the GPU-only approach. Further, to scale HOSTA on TianHe-1A, we propose a gather/scatter optimization to minimize PCI-e data transfer times for ghost and singularity data of 3D grid blocks, and overlap the collaborative computation and communication as far as possible using some advanced CUDA and MPI features. Scalability tests show that HOSTA can achieve a parallel efficiency of above 60% on 1024 TianHe-1A nodes. With our method, we have successfully simulated an EET high-lift airfoil configuration containing 800M cells and China's large civil airplane configuration containing 150M cells. To our best knowledge, those are the largest-scale CPU–GPU collaborative simulations

The GF11 supercomputer

International Nuclear Information System (INIS)

Beetem, J.; Weingarten, D.

1986-01-01

GF11 is a parallel computer currently under construction at the IBM Yorktown Research Center. The machine incorporates 576 floating-point processors arrangedin a modified SIMD architecture. Each has space for 2 Mbytes of memory and is capable of 20 Mflops, giving the total machine a peak of 1.125 Gbytes of memory and 11.52 Gflops. The floating-point processors are interconnected by a dynamically reconfigurable non-blocking switching network. At each machine cycle any of 1024 pre-selected permutations of data can be realized among the processors. The main intended application of GF11 is a class of calculations arising from quantum chromodynamics

The GF11 supercomputer

International Nuclear Information System (INIS)

Beetem, J.; Denneau, M.; Weingarten, D.

1985-01-01

GF11 is a parallel computer currently under construction at the IBM Yorktown Research Center. The machine incorporates 576 floating- point processors arranged in a modified SIMD architecture. Each has space for 2 Mbytes of memory and is capable of 20 Mflops, giving the total machine a peak of 1.125 Gbytes of memory and 11.52 Gflops. The floating-point processors are interconnected by a dynamically reconfigurable nonblocking switching network. At each machine cycle any of 1024 pre-selected permutations of data can be realized among the processors. The main intended application of GF11 is a class of calculations arising from quantum chromodynamics

Supercomputer debugging workshop `92

Energy Technology Data Exchange (ETDEWEB)

Brown, J.S.

1993-02-01

This report contains papers or viewgraphs on the following topics: The ABCs of Debugging in the 1990s; Cray Computer Corporation; Thinking Machines Corporation; Cray Research, Incorporated; Sun Microsystems, Inc; Kendall Square Research; The Effects of Register Allocation and Instruction Scheduling on Symbolic Debugging; Debugging Optimized Code: Currency Determination with Data Flow; A Debugging Tool for Parallel and Distributed Programs; Analyzing Traces of Parallel Programs Containing Semaphore Synchronization; Compile-time Support for Efficient Data Race Detection in Shared-Memory Parallel Programs; Direct Manipulation Techniques for Parallel Debuggers; Transparent Observation of XENOOPS Objects; A Parallel Software Monitor for Debugging and Performance Tools on Distributed Memory Multicomputers; Profiling Performance of Inter-Processor Communications in an iWarp Torus; The Application of Code Instrumentation Technology in the Los Alamos Debugger; and CXdb: The Road to Remote Debugging.

The GF11 supercomputer

International Nuclear Information System (INIS)

Beetem, J.; Denneau, M.; Weingarten, D.

1985-01-01

GF11 is a parallel computer currently under construction at the Yorktown Research Center. The machine incorporates 576 floating-point processors arranged in a modified SIMD architecture. Each processor has space for 2 Mbytes of memory and is capable of 20 MFLOPS, giving the total machine a peak of 1.125 Gbytes of memory and 11.52 GFLOPS. The floating-point processors are interconnected by a dynamically reconfigurable non-blocking switching network. At each machine cycle any of 1024 pre-selected permutations of data can be realized among the processors. The main intended application of GF11 is a class of calculations arising from quantum chromodynamics, a proposed theory of the elementary particles which participate in nuclear interactions

Easy Access to HPC Resources through the Application GUI

KAUST Repository

van Waveren, Matthijs

2016-11-01

The computing environment at the King Abdullah University of Science and Technology (KAUST) is growing in size and complexity. KAUST hosts the tenth fastest supercomputer in the world (Shaheen II) and several HPC clusters. Researchers can be inhibited by the complexity, as they need to learn new languages and execute many tasks in order to access the HPC clusters and the supercomputer. In order to simplify the access, we have developed an interface between the applications and the clusters and supercomputer that automates the transfer of input data and job submission and also the retrieval of results to the researcher’s local workstation. The innovation is that the user now submits his jobs from within the application GUI on his workstation, and does not have to directly log into the clusters or supercomputer anymore. This article details the solution and its benefits to the researchers.

Parallel processor programs in the Federal Government

Science.gov (United States)

Schneck, P. B.; Austin, D.; Squires, S. L.; Lehmann, J.; Mizell, D.; Wallgren, K.

1985-01-01

In 1982, a report dealing with the nation's research needs in high-speed computing called for increased access to supercomputing resources for the research community, research in computational mathematics, and increased research in the technology base needed for the next generation of supercomputers. Since that time a number of programs addressing future generations of computers, particularly parallel processors, have been started by U.S. government agencies. The present paper provides a description of the largest government programs in parallel processing. Established in fiscal year 1985 by the Institute for Defense Analyses for the National Security Agency, the Supercomputing Research Center will pursue research to advance the state of the art in supercomputing. Attention is also given to the DOE applied mathematical sciences research program, the NYU Ultracomputer project, the DARPA multiprocessor system architectures program, NSF research on multiprocessor systems, ONR activities in parallel computing, and NASA parallel processor projects.

Building the Teraflops/Petabytes Production Computing Center

International Nuclear Information System (INIS)

Kramer, William T.C.; Lucas, Don; Simon, Horst D.

1999-01-01

In just one decade, the 1990s, supercomputer centers have undergone two fundamental transitions which require rethinking their operation and their role in high performance computing. The first transition in the early to mid-1990s resulted from a technology change in high performance computing architecture. Highly parallel distributed memory machines built from commodity parts increased the operational complexity of the supercomputer center, and required the introduction of intellectual services as equally important components of the center. The second transition is happening in the late 1990s as centers are introducing loosely coupled clusters of SMPs as their premier high performance computing platforms, while dealing with an ever-increasing volume of data. In addition, increasing network bandwidth enables new modes of use of a supercomputer center, in particular, computational grid applications. In this paper we describe what steps NERSC is taking to address these issues and stay at the leading edge of supercomputing centers.; N

Exascale Data Analysis

CERN Multimedia

CERN. Geneva; Fitch, Blake

2011-01-01

Traditionaly, the primary role of supercomputers was to create data, primarily for simulation applications. Due to usage and technology trends, supercomputers are increasingly also used for data analysis. Some of this data is from simulations, but there is also a rapidly increasingly amount of real-world science and business data to be analyzed. We briefly overview Blue Gene and other current supercomputer architectures. We outline future architectures, up to the Exascale supercomputers expected in the 2020 time frame. We focus on the data analysis challenges and opportunites, especially those concerning Flash and other up-and-coming storage class memory. About the speakers Blake G. Fitch has been with IBM Research, Yorktown Heights, NY since 1987, mainly pursuing interests in parallel systems. He joined the Scalable Parallel Systems Group in 1990, contributing to research and development that culminated in the IBM scalable parallel system (SP*) product. His research interests have focused on applicatio...

Lattice gauge theory using parallel processors

International Nuclear Information System (INIS)

Lee, T.D.; Chou, K.C.; Zichichi, A.

1987-01-01

The book's contents include: Lattice Gauge Theory Lectures: Introduction and Current Fermion Simulations; Monte Carlo Algorithms for Lattice Gauge Theory; Specialized Computers for Lattice Gauge Theory; Lattice Gauge Theory at Finite Temperature: A Monte Carlo Study; Computational Method - An Elementary Introduction to the Langevin Equation, Present Status of Numerical Quantum Chromodynamics; Random Lattice Field Theory; The GF11 Processor and Compiler; and The APE Computer and First Physics Results; Columbia Supercomputer Project: Parallel Supercomputer for Lattice QCD; Statistical and Systematic Errors in Numerical Simulations; Monte Carlo Simulation for LGT and Programming Techniques on the Columbia Supercomputer; Food for Thought: Five Lectures on Lattice Gauge Theory

Advances in petascale kinetic plasma simulation with VPIC and Roadrunner

Energy Technology Data Exchange (ETDEWEB)

Bowers, Kevin J [Los Alamos National Laboratory; Albright, Brian J [Los Alamos National Laboratory; Yin, Lin [Los Alamos National Laboratory; Daughton, William S [Los Alamos National Laboratory; Roytershteyn, Vadim [Los Alamos National Laboratory; Kwan, Thomas J T [Los Alamos National Laboratory

2009-01-01

VPIC, a first-principles 3d electromagnetic charge-conserving relativistic kinetic particle-in-cell (PIC) code, was recently adapted to run on Los Alamos's Roadrunner, the first supercomputer to break a petaflop (10{sup 15} floating point operations per second) in the TOP500 supercomputer performance rankings. They give a brief overview of the modeling capabilities and optimization techniques used in VPIC and the computational characteristics of petascale supercomputers like Roadrunner. They then discuss three applications enabled by VPIC's unprecedented performance on Roadrunner: modeling laser plasma interaction in upcoming inertial confinement fusion experiments at the National Ignition Facility (NIF), modeling short pulse laser GeV ion acceleration and modeling reconnection in magnetic confinement fusion experiments.

Lawrence Livermore National Laboratory selects Intel Itanium 2 processors for world's most powerful Linux cluster

CERN Multimedia

2003-01-01

"Intel Corporation, system manufacturer California Digital and the University of California at Lawrence Livermore National Laboratory (LLNL) today announced they are building one of the world's most powerful supercomputers. The supercomputer project, codenamed "Thunder," uses nearly 4,000 Intel® Itanium® 2 processors... is expected to be complete in January 2004" (1 page).

Parallel adaptation of a vectorised quantumchemical program system

International Nuclear Information System (INIS)

Van Corler, L.C.H.; Van Lenthe, J.H.

1987-01-01

Supercomputers, like the CRAY 1 or the Cyber 205, have had, and still have, a marked influence on Quantum Chemistry. Vectorization has led to a considerable increase in the performance of Quantum Chemistry programs. However, clockcycle times more than a factor 10 smaller than those of the present supercomputers are not to be expected. Therefore future supercomputers will have to depend on parallel structures. Recently, the first examples of such supercomputers have been installed. To be prepared for this new generation of (parallel) supercomputers one should consider the concepts one wants to use and the kind of problems one will encounter during implementation of existing vectorized programs on those parallel systems. The authors implemented four important parts of a large quantumchemical program system (ATMOL), i.e. integrals, SCF, 4-index and Direct-CI in the parallel environment at ECSEC (Rome, Italy). This system offers simulated parallellism on the host computer (IBM 4381) and real parallellism on at most 10 attached processors (FPS-164). Quantumchemical programs usually handle large amounts of data and very large, often sparse matrices. The transfer of that many data can cause problems concerning communication and overhead, in view of which shared memory and shared disks must be considered. The strategy and the tools that were used to parallellise the programs are shown. Also, some examples are presented to illustrate effectiveness and performance of the system in Rome for these type of calculations

Evaluation of existing and proposed computer architectures for future ground-based systems

Science.gov (United States)

Schulbach, C.

1985-01-01

Parallel processing architectures and techniques used in current supercomputers are described and projections are made of future advances. Presently, the von Neumann sequential processing pattern has been accelerated by having separate I/O processors, interleaved memories, wide memories, independent functional units and pipelining. Recent supercomputers have featured single-input, multiple data stream architectures, which have different processors for performing various operations (vector or pipeline processors). Multiple input, multiple data stream machines have also been developed. Data flow techniques, wherein program instructions are activated only when data are available, are expected to play a large role in future supercomputers, along with increased parallel processor arrays. The enhanced operational speeds are essential for adequately treating data from future spacecraft remote sensing instruments such as the Thematic Mapper.

[Teacher enhancement at Supercomputing `96

Energy Technology Data Exchange (ETDEWEB)

NONE

1998-02-13

The SC`96 Education Program provided a three-day professional development experience for middle and high school science, mathematics, and computer technology teachers. The program theme was Computers at Work in the Classroom, and a majority of the sessions were presented by classroom teachers who have had several years experience in using these technologies with their students. The teachers who attended the program were introduced to classroom applications of computing and networking technologies and were provided to the greatest extent possible with lesson plans, sample problems, and other resources that could immediately be used in their own classrooms. The attached At a Glance Schedule and Session Abstracts describes in detail the three-day SC`96 Education Program. Also included is the SC`96 Education Program evaluation report and the financial report.

Mantle Convection on Modern Supercomputers

Science.gov (United States)

Weismüller, J.; Gmeiner, B.; Huber, M.; John, L.; Mohr, M.; Rüde, U.; Wohlmuth, B.; Bunge, H. P.

2015-12-01

Mantle convection is the cause for plate tectonics, the formation of mountains and oceans, and the main driving mechanism behind earthquakes. The convection process is modeled by a system of partial differential equations describing the conservation of mass, momentum and energy. Characteristic to mantle flow is the vast disparity of length scales from global to microscopic, turning mantle convection simulations into a challenging application for high-performance computing. As system size and technical complexity of the simulations continue to increase, design and implementation of simulation models for next generation large-scale architectures is handled successfully only in an interdisciplinary context. A new priority program - named SPPEXA - by the German Research Foundation (DFG) addresses this issue, and brings together computer scientists, mathematicians and application scientists around grand challenges in HPC. Here we report from the TERRA-NEO project, which is part of the high visibility SPPEXA program, and a joint effort of four research groups. TERRA-NEO develops algorithms for future HPC infrastructures, focusing on high computational efficiency and resilience in next generation mantle convection models. We present software that can resolve the Earth's mantle with up to 1012 grid points and scales efficiently to massively parallel hardware with more than 50,000 processors. We use our simulations to explore the dynamic regime of mantle convection and assess the impact of small scale processes on global mantle flow.

Advanced Architectures for Astrophysical Supercomputing

Science.gov (United States)

Barsdell, B. R.; Barnes, D. G.; Fluke, C. J.

2010-12-01

Astronomers have come to rely on the increasing performance of computers to reduce, analyze, simulate and visualize their data. In this environment, faster computation can mean more science outcomes or the opening up of new parameter spaces for investigation. If we are to avoid major issues when implementing codes on advanced architectures, it is important that we have a solid understanding of our algorithms. A recent addition to the high-performance computing scene that highlights this point is the graphics processing unit (GPU). The hardware originally designed for speeding-up graphics rendering in video games is now achieving speed-ups of O(100×) in general-purpose computation - performance that cannot be ignored. We are using a generalized approach, based on the analysis of astronomy algorithms, to identify the optimal problem-types and techniques for taking advantage of both current GPU hardware and future developments in computing architectures.

Supercomputer debugging workshop '92

Energy Technology Data Exchange (ETDEWEB)

Brown, J.S.

1993-01-01

This report contains papers or viewgraphs on the following topics: The ABCs of Debugging in the 1990s; Cray Computer Corporation; Thinking Machines Corporation; Cray Research, Incorporated; Sun Microsystems, Inc; Kendall Square Research; The Effects of Register Allocation and Instruction Scheduling on Symbolic Debugging; Debugging Optimized Code: Currency Determination with Data Flow; A Debugging Tool for Parallel and Distributed Programs; Analyzing Traces of Parallel Programs Containing Semaphore Synchronization; Compile-time Support for Efficient Data Race Detection in Shared-Memory Parallel Programs; Direct Manipulation Techniques for Parallel Debuggers; Transparent Observation of XENOOPS Objects; A Parallel Software Monitor for Debugging and Performance Tools on Distributed Memory Multicomputers; Profiling Performance of Inter-Processor Communications in an iWarp Torus; The Application of Code Instrumentation Technology in the Los Alamos Debugger; and CXdb: The Road to Remote Debugging.

ASCI's Vision for supercomputing future

International Nuclear Information System (INIS)

Nowak, N.D.

2003-01-01

The full text of publication follows. Advanced Simulation and Computing (ASC, formerly Accelerated Strategic Computing Initiative [ASCI]) was established in 1995 to help Defense Programs shift from test-based confidence to simulation-based confidence. Specifically, ASC is a focused and balanced program that is accelerating the development of simulation capabilities needed to analyze and predict the performance, safety, and reliability of nuclear weapons and certify their functionality - far exceeding what might have been achieved in the absence of a focused initiative. To realize its vision, ASC is creating simulation and proto-typing capabilities, based on advanced weapon codes and high-performance computing

Supercomputer requirements for theoretical chemistry

International Nuclear Information System (INIS)

Walker, R.B.; Hay, P.J.; Galbraith, H.W.

1980-01-01

Many problems important to the theoretical chemist would, if implemented in their full complexity, strain the capabilities of today's most powerful computers. Several such problems are now being implemented on the CRAY-1 computer at Los Alamos. Examples of these problems are taken from the fields of molecular electronic structure calculations, quantum reactive scattering calculations, and quantum optics. 12 figures

A portable grid-enabled computing system for a nuclear material study

International Nuclear Information System (INIS)

Tsujita, Yuichi; Arima, Tatsumi; Takekawa, Takayuki; Suzuki, Yoshio

2010-01-01

We have built a portable grid-enabled computing system specialized for our molecular dynamics (MD) simulation program to study Pu material easily. Experimental approach to reveal properties of Pu materials is often accompanied by some difficulties such as radiotoxicity of actinides. Since a computational approach reveals new aspects to researchers without such radioactive facilities, we address an MD computation. In order to have more realistic results about e.g., melting point or thermal conductivity, we need a large scale of parallel computations. Most of application users who don't have supercomputers in their institutes should use a remote supercomputer. For such users, we have developed the portable and secured grid-enabled computing system to utilize a grid computing infrastructure provided by Information Technology Based Laboratory (ITBL). This system enables us to access remote supercomputers in the ITBL system seamlessly from a client PC through its graphical user interface (GUI). Typically it enables seamless file accesses on the GUI. Furthermore monitoring of standard output or standard error is available to see progress of an executed program. Since the system provides fruitful functionalities which are useful for parallel computing on a remote supercomputer, application users can concentrate on their researches. (author)

Development of in-situ visualization tool for PIC simulation

International Nuclear Information System (INIS)

Ohno, Nobuaki; Ohtani, Hiroaki

2014-01-01

As the capability of a supercomputer is improved, the sizes of simulation and its output data also become larger and larger. Visualization is usually carried out on a researcher's PC with interactive visualization software after performing the computer simulation. However, the data size is becoming too large to do it currently. A promising answer is in-situ visualization. For this case a simulation code is coupled with the visualization code and visualization is performed with the simulation on the same supercomputer. We developed an in-situ visualization tool for particle-in-cell (PIC) simulation and it is provided as a Fortran's module. We coupled it with a PIC simulation code and tested the coupled code on Plasma Simulator supercomputer, and ensured that it works. (author)

Aviation Research and the Internet

Science.gov (United States)

Scott, Antoinette M.

1995-01-01

The Internet is a network of networks. It was originally funded by the Defense Advanced Research Projects Agency or DOD/DARPA and evolved in part from the connection of supercomputer sites across the United States. The National Science Foundation (NSF) made the most of their supercomputers by connecting the sites to each other. This made the supercomputers more efficient and now allows scientists, engineers and researchers to access the supercomputers from their own labs and offices. The high speed networks that connect the NSF supercomputers form the backbone of the Internet. The World Wide Web (WWW) is a menu system. It gathers Internet resources from all over the world into a series of screens that appear on your computer. The WWW is also a distributed. The distributed system stores data information on many computers (servers). These servers can go out and get data when you ask for it. Hypermedia is the base of the WWW. One can 'click' on a section and visit other hypermedia (pages). Our approach to demonstrating the importance of aviation research through the Internet began with learning how to put pages on the Internet (on-line) ourselves. We were assigned two aviation companies; Vision Micro Systems Inc. and Innovative Aerodynamic Technologies (IAT). We developed home pages for these SBIR companies. The equipment used to create the pages were the UNIX and Macintosh machines. HTML Supertext software was used to write the pages and the Sharp JX600S scanner to scan the images. As a result, with the use of the UNIX, Macintosh, Sun, PC, and AXIL machines, we were able to present our home pages to over 800,000 visitors.

Fast methods for long-range interactions in complex systems. Lecture notes

Energy Technology Data Exchange (ETDEWEB)

Sutmann, Godehard; Gibbon, Paul; Lippert, Thomas (eds.)

2011-10-13

Parallel computing and computer simulations of complex particle systems including charges have an ever increasing impact in a broad range of fields in the physical sciences, e.g. in astrophysics, statistical physics, plasma physics, material sciences, physical chemistry, and biophysics. The present summer school, funded by the German Heraeus-Foundation, took place at the Juelich Supercomputing Centre from 6 - 10 September 2010. The focus was on providing an introduction and overview over different methods, algorithms and new trends for the computational treatment of long-range interactions in particle systems. The Lecture Notes contain an introduction into particle simulation, as well as five different fast methods, i.e. the Fast Multipole Method, Barnes-Hut Tree Method, Multigrid, FFT based methods, and Fast Summation using the non-equidistant FFT. In addition to introducing the methods, efficient parallelization of the methods is presented in detail. This publication was edited at the Juelich Supercomputing Centre (JSC) which is an integral part of the Institute for Advanced Simulation (IAS). The IAS combines the Juelich simulation sciences and the supercomputer facility in one organizational unit. It includes those parts of the scientific institutes at Forschungszentrum Juelich which use simulation on supercomputers as their main research methodology. (orig.)

Fast methods for long-range interactions in complex systems. Lecture notes

International Nuclear Information System (INIS)

Sutmann, Godehard; Gibbon, Paul; Lippert, Thomas

2011-01-01

Parallel computing and computer simulations of complex particle systems including charges have an ever increasing impact in a broad range of fields in the physical sciences, e.g. in astrophysics, statistical physics, plasma physics, material sciences, physical chemistry, and biophysics. The present summer school, funded by the German Heraeus-Foundation, took place at the Juelich Supercomputing Centre from 6 - 10 September 2010. The focus was on providing an introduction and overview over different methods, algorithms and new trends for the computational treatment of long-range interactions in particle systems. The Lecture Notes contain an introduction into particle simulation, as well as five different fast methods, i.e. the Fast Multipole Method, Barnes-Hut Tree Method, Multigrid, FFT based methods, and Fast Summation using the non-equidistant FFT. In addition to introducing the methods, efficient parallelization of the methods is presented in detail. This publication was edited at the Juelich Supercomputing Centre (JSC) which is an integral part of the Institute for Advanced Simulation (IAS). The IAS combines the Juelich simulation sciences and the supercomputer facility in one organizational unit. It includes those parts of the scientific institutes at Forschungszentrum Juelich which use simulation on supercomputers as their main research methodology. (orig.)

Unified, Cross-Platform, Open-Source Library Package for High-Performance Computing

Energy Technology Data Exchange (ETDEWEB)

Kozacik, Stephen [EM Photonics, Inc., Newark, DE (United States)

2017-05-15

Compute power is continually increasing, but this increased performance is largely found in sophisticated computing devices and supercomputer resources that are difficult to use, resulting in under-utilization. We developed a unified set of programming tools that will allow users to take full advantage of the new technology by allowing them to work at a level abstracted away from the platform specifics, encouraging the use of modern computing systems, including government-funded supercomputer facilities.

Algorithm comparison and benchmarking using a parallel spectra transform shallow water model

Energy Technology Data Exchange (ETDEWEB)

Worley, P.H. [Oak Ridge National Lab., TN (United States); Foster, I.T.; Toonen, B. [Argonne National Lab., IL (United States)

1995-04-01

In recent years, a number of computer vendors have produced supercomputers based on a massively parallel processing (MPP) architecture. These computers have been shown to be competitive in performance with conventional vector supercomputers for some applications. As spectral weather and climate models are heavy users of vector supercomputers, it is interesting to determine how these models perform on MPPS, and which MPPs are best suited to the execution of spectral models. The benchmarking of MPPs is complicated by the fact that different algorithms may be more efficient on different architectures. Hence, a comprehensive benchmarking effort must answer two related questions: which algorithm is most efficient on each computer and how do the most efficient algorithms compare on different computers. In general, these are difficult questions to answer because of the high cost associated with implementing and evaluating a range of different parallel algorithms on each MPP platform.

Parallel-Vector Algorithm For Rapid Structural Anlysis

Science.gov (United States)

Agarwal, Tarun R.; Nguyen, Duc T.; Storaasli, Olaf O.

1993-01-01

New algorithm developed to overcome deficiency of skyline storage scheme by use of variable-band storage scheme. Exploits both parallel and vector capabilities of modern high-performance computers. Gives engineers and designers opportunity to include more design variables and constraints during optimization of structures. Enables use of more refined finite-element meshes to obtain improved understanding of complex behaviors of aerospace structures leading to better, safer designs. Not only attractive for current supercomputers but also for next generation of shared-memory supercomputers.

Scalability of DL_POLY on High Performance Computing Platform

CSIR Research Space (South Africa)

Mabakane, Mabule S

2017-12-01

Full Text Available stream_source_info Mabakanea_19979_2017.pdf.txt stream_content_type text/plain stream_size 33716 Content-Encoding UTF-8 stream_name Mabakanea_19979_2017.pdf.txt Content-Type text/plain; charset=UTF-8 SACJ 29(3) December... when using many processors within the compute nodes of the supercomputer. The type of the processors of compute nodes and their memory also play an important role in the overall performance of the parallel application running on a supercomputer. DL...

Strategic research field no.4, industrial innovations

International Nuclear Information System (INIS)

Kato, Chisachi

2011-01-01

'Kei'-supercomputer is planned to start its full-scale operation in about one year and a half. With this, High Performance Computing (HPC) is most likely to contribute not only to further progress in basic and applied sciences, but also to bringing about innovations in various fields of industries. It is expected to substantially shorten design time, drastically improve performance and/or liability of various industrial products, and greatly enhance safety of large-scale power plants. In this particle, six research themes, which are currently being prepared in this strategic research field, 'industrial innovations' so as to use 'Kei'-supercomputer as soon as it starts operations, will be briefly described regarding their specific goals and break-through that they are expected to bring about in industries. It is also explained how we have determined these themes. We are also planning several measures in order to promote widespread use of HPC including 'Kei'-supercomputer in industries, which will also be elaborated in this article. (author)

Distributed interactive graphics applications in computational fluid dynamics

International Nuclear Information System (INIS)

Rogers, S.E.; Buning, P.G.; Merritt, F.J.

1987-01-01

Implementation of two distributed graphics programs used in computational fluid dynamics is discussed. Both programs are interactive in nature. They run on a CRAY-2 supercomputer and use a Silicon Graphics Iris workstation as the front-end machine. The hardware and supporting software are from the Numerical Aerodynamic Simulation project. The supercomputer does all numerically intensive work and the workstation, as the front-end machine, allows the user to perform real-time interactive transformations on the displayed data. The first program was written as a distributed program that computes particle traces for fluid flow solutions existing on the supercomputer. The second is an older post-processing and plotting program modified to run in a distributed mode. Both programs have realized a large increase in speed over that obtained using a single machine. By using these programs, one can learn quickly about complex features of a three-dimensional flow field. Some color results are presented

An evaluation of current high-performance networks

Energy Technology Data Exchange (ETDEWEB)

Bell, Christian; Bonachea, Dan; Cote, Yannick; Duell, Jason; Hargrove, Paul; Husbands, Parry; Iancu, Costin; Welcome, Michael; Yelick, Katherine

2003-01-25

High-end supercomputers are increasingly built out of commodity components, and lack tight integration between the processor and network. This often results in inefficiencies in the communication subsystem, such as high software overheads and/or message latencies. In this paper we use a set of microbenchmarks to quantify the cost of this commoditization, measuring software overhead, latency, and bandwidth on five contemporary supercomputing networks. We compare the performance of the ubiquitous MPI layer to that of lower-level communication layers, and quantify the advantages of the latter for small message performance. We also provide data on the potential for various communication-related optimizations, such as overlapping communication with computation or other communication. Finally, we determine the minimum size needed for a message to be considered 'large' (i.e., bandwidth-bound) on these platforms, and provide historical data on the software overheads of a number of supercomputers over the past decade.

Supercomputer modeling of volcanic eruption dynamics

Energy Technology Data Exchange (ETDEWEB)

Kieffer, S.W. [Arizona State Univ., Tempe, AZ (United States); Valentine, G.A. [Los Alamos National Lab., NM (United States); Woo, Mahn-Ling [Arizona State Univ., Tempe, AZ (United States)

1995-06-01

Our specific goals are to: (1) provide a set of models based on well-defined assumptions about initial and boundary conditions to constrain interpretations of observations of active volcanic eruptions--including movies of flow front velocities, satellite observations of temperature in plumes vs. time, and still photographs of the dimensions of erupting plumes and flows on Earth and other planets; (2) to examine the influence of subsurface conditions on exit plane conditions and plume characteristics, and to compare the models of subsurface fluid flow with seismic constraints where possible; (3) to relate equations-of-state for magma-gas mixtures to flow dynamics; (4) to examine, in some detail, the interaction of the flowing fluid with the conduit walls and ground topography through boundary layer theory so that field observations of erosion and deposition can be related to fluid processes; and (5) to test the applicability of existing two-phase flow codes for problems related to the generation of volcanic long-period seismic signals; (6) to extend our understanding and simulation capability to problems associated with emplacement of fragmental ejecta from large meteorite impacts.

Trends in supercomputers and computational physics

International Nuclear Information System (INIS)

Bloch, T.

1985-01-01

Today, scientists using numerical models explore the basic mechanisms of semiconductors, apply global circulation models to climatic and oceanographic problems, probe into the behaviour of galaxies and try to verify basic theories of matter, such as Quantum Chromo Dynamics by simulating the constitution of elementary particles. Chemists, crystallographers and molecular dynamics researchers develop models for chemical reactions, formation of crystals and try to deduce the chemical properties of molecules as a function of the shapes of their states. Chaotic systems are studied extensively in turbulence (combustion included) and the design of the next generation of controlled fusion devices relies heavily on computational physics. (orig./HSI)

A supercomputer for parallel data analysis

International Nuclear Information System (INIS)

Kolpakov, I.F.; Senner, A.E.; Smirnov, V.A.

1987-01-01

The project of a powerful multiprocessor system is proposed. The main purpose of the project is to develop a low cost computer system with a processing rate of a few tens of millions of operations per second. The system solves many problems of data analysis from high-energy physics spectrometers. It includes about 70 MOTOROLA-68020 based powerful slave microprocessor boards liaisoned through the VME crates to a host VAX micro computer. Each single microprocessor board performs the same algorithm requiring large computing time. The host computer distributes data over the microprocessor board, collects and combines obtained results. The architecture of the system easily allows one to use it in the real time mode

Large scale visualization on the Cray XT3 using ParaView.

Energy Technology Data Exchange (ETDEWEB)

Rogers, David; Geveci, Berk (Kitware, Inc.); Eschenbert, Kent (Pittsburgh Supercomputing Center); Neundorf, Alexander (Technical University of Kaiserslautern); Marion, Patrick (Kitware, Inc.); Moreland, Kenneth D.; Greenfield, John

2008-05-01

Post-processing and visualization are key components to understanding any simulation. Porting ParaView, a scalable visualization tool, to the Cray XT3 allows our analysts to leverage the same supercomputer they use for simulation to perform post-processing. Visualization tools traditionally rely on a variety of rendering, scripting, and networking resources; the challenge of running ParaView on the Lightweight Kernel is to provide and use the visualization and post-processing features in the absence of many OS resources. We have successfully accomplished this at Sandia National Laboratories and the Pittsburgh Supercomputing Center.

Magnetic fusion energy and computers. The role of computing in magnetic fusion energy research and development (second edition)

International Nuclear Information System (INIS)

1983-01-01

This report documents the structure and uses of the MFE Network and presents a compilation of future computing requirements. Its primary emphasis is on the role of supercomputers in fusion research. One of its key findings is that with the introduction of each successive class of supercomputer, qualitatively improved understanding of fusion processes has been gained. At the same time, even the current Class VI machines severely limit the attainable realism of computer models. Many important problems will require the introduction of Class VII or even larger machines before they can be successfully attacked

Publisher Correction

DEFF Research Database (Denmark)

Bonàs-Guarch, Sílvia; Guindo-Martínez, Marta; Miguel-Escalada, Irene

2018-01-01

In the originally published version of this Article, the affiliation details for Santi González, Jian'an Luan and Claudia Langenberg were inadvertently omitted. Santi González should have been affiliated with 'Barcelona Supercomputing Center (BSC), Joint BSC-CRG-IRB Research Program in Computatio......In the originally published version of this Article, the affiliation details for Santi González, Jian'an Luan and Claudia Langenberg were inadvertently omitted. Santi González should have been affiliated with 'Barcelona Supercomputing Center (BSC), Joint BSC-CRG-IRB Research Program...

Challenges in scaling NLO generators to leadership computers

Science.gov (United States)

Benjamin, D.; Childers, JT; Hoeche, S.; LeCompte, T.; Uram, T.

2017-10-01

Exascale computing resources are roughly a decade away and will be capable of 100 times more computing than current supercomputers. In the last year, Energy Frontier experiments crossed a milestone of 100 million core-hours used at the Argonne Leadership Computing Facility, Oak Ridge Leadership Computing Facility, and NERSC. The Fortran-based leading-order parton generator called Alpgen was successfully scaled to millions of threads to achieve this level of usage on Mira. Sherpa and MadGraph are next-to-leading order generators used heavily by LHC experiments for simulation. Integration times for high-multiplicity or rare processes can take a week or more on standard Grid machines, even using all 16-cores. We will describe our ongoing work to scale the Sherpa generator to thousands of threads on leadership-class machines and reduce run-times to less than a day. This work allows the experiments to leverage large-scale parallel supercomputers for event generation today, freeing tens of millions of grid hours for other work, and paving the way for future applications (simulation, reconstruction) on these and future supercomputers.

Application experiences with the Globus toolkit.

Energy Technology Data Exchange (ETDEWEB)

Brunett, S.

1998-06-09

The Globus grid toolkit is a collection of software components designed to support the development of applications for high-performance distributed computing environments, or ''computational grids'' [14]. The Globus toolkit is an implementation of a ''bag of services'' architecture, which provides application and tool developers not with a monolithic system but rather with a set of stand-alone services. Each Globus component provides a basic service, such as authentication, resource allocation, information, communication, fault detection, and remote data access. Different applications and tools can combine these services in different ways to construct ''grid-enabled'' systems. The Globus toolkit has been used to construct the Globus Ubiquitous Supercomputing Testbed, or GUSTO: a large-scale testbed spanning 20 sites and included over 4000 compute nodes for a total compute power of over 2 TFLOPS. Over the past six months, we and others have used this testbed to conduct a variety of application experiments, including multi-user collaborative environments (tele-immersion), computational steering, distributed supercomputing, and high throughput computing. The goal of this paper is to review what has been learned from these experiments regarding the effectiveness of the toolkit approach. To this end, we describe two of the application experiments in detail, noting what worked well and what worked less well. The two applications are a distributed supercomputing application, SF-Express, in which multiple supercomputers are harnessed to perform large distributed interactive simulations; and a tele-immersion application, CAVERNsoft, in which the focus is on connecting multiple people to a distributed simulated world.

Productive Parallel Programming: The PCN Approach

Directory of Open Access Journals (Sweden)

Ian Foster

1992-01-01

Full Text Available We describe the PCN programming system, focusing on those features designed to improve the productivity of scientists and engineers using parallel supercomputers. These features include a simple notation for the concise specification of concurrent algorithms, the ability to incorporate existing Fortran and C code into parallel applications, facilities for reusing parallel program components, a portable toolkit that allows applications to be developed on a workstation or small parallel computer and run unchanged on supercomputers, and integrated debugging and performance analysis tools. We survey representative scientific applications and identify problem classes for which PCN has proved particularly useful.

Status of the Fermilab lattice supercomputer project

International Nuclear Information System (INIS)

Mackenzie, P.; Eichten, E.; Hockney, G.

1988-10-01

Fermilab has completed construction of a sixteen node (320 megaflop peak speed) parallel computer for lattice gauge theory calculations. The architecture was designed to provide the highest possible cost effectiveness while maintaining a high level of programmability and constraining as little as possible the types of lattice problems which can be done on it. The machine is programmed in C. It is a prototype for a 256 node (5 gigaflop peak speed) computer which will be assembled this winter. 6 refs

Experimental HEP supercomputing at F.S.U

International Nuclear Information System (INIS)

Levinthal, D.; Goldman, H.; Hodous, M.F.

1987-01-01

We have developed a track reconstruction algorithm that will work with any 2-dimensional detector as long as the 2-dimensional projections of that detector are ''true'' projections. The program is implemented on a 2-pipe CDC CYBER-205 with 4 million words of memory. (orig./HSI)

Data Mining Supercomputing with SAS JMP® Genomics

Directory of Open Access Journals (Sweden)

Richard S. Segall

2011-02-01

Full Text Available JMP® Genomics is statistical discovery software that can uncover meaningful patterns in high-throughput genomics and proteomics data. JMP® Genomics is designed for biologists, biostatisticians, statistical geneticists, and those engaged in analyzing the vast stores of data that are common in genomic research (SAS, 2009. Data mining was performed using JMP® Genomics on the two collections of microarray databases available from National Center for Biotechnology Information (NCBI for lung cancer and breast cancer. The Gene Expression Omnibus (GEO of NCBI serves as a public repository for a wide range of highthroughput experimental data, including the two collections of lung cancer and breast cancer that were used for this research. The results for applying data mining using software JMP® Genomics are shown in this paper with numerous screen shots.

Performance Analysis of FEM Algorithmson GPU and Many-Core Architectures

KAUST Repository

Khurram, Rooh

2015-04-27

The roadmaps of the leading supercomputer manufacturers are based on hybrid systems, which consist of a mix of conventional processors and accelerators. This trend is mainly due to the fact that the power consumption cost of the future cpu-only Exascale systems will be unsustainable, thus accelerators such as graphic processing units (GPUs) and many-integrated-core (MIC) will likely be the integral part of the TOP500 (http://www.top500.org/) supercomputers, beyond 2020. The emerging supercomputer architecture will bring new challenges for the code developers. Continuum mechanics codes will particularly be affected, because the traditional synchronous implicit solvers will probably not scale on hybrid Exascale machines. In the previous study[1], we reported on the performance of a conjugate gradient based mesh motion algorithm[2]on Sandy Bridge, Xeon Phi, and K20c. In the present study we report on the comparative study of finite element codes, using PETSC and AmgX solvers on CPU and GPUs, respectively [3,4]. We believe this study will be a good starting point for FEM code developers, who are contemplating a CPU to accelerator transition.

SiGN-SSM: open source parallel software for estimating gene networks with state space models.

Science.gov (United States)

Tamada, Yoshinori; Yamaguchi, Rui; Imoto, Seiya; Hirose, Osamu; Yoshida, Ryo; Nagasaki, Masao; Miyano, Satoru

2011-04-15

SiGN-SSM is an open-source gene network estimation software able to run in parallel on PCs and massively parallel supercomputers. The software estimates a state space model (SSM), that is a statistical dynamic model suitable for analyzing short time and/or replicated time series gene expression profiles. SiGN-SSM implements a novel parameter constraint effective to stabilize the estimated models. Also, by using a supercomputer, it is able to determine the gene network structure by a statistical permutation test in a practical time. SiGN-SSM is applicable not only to analyzing temporal regulatory dependencies between genes, but also to extracting the differentially regulated genes from time series expression profiles. SiGN-SSM is distributed under GNU Affero General Public Licence (GNU AGPL) version 3 and can be downloaded at http://sign.hgc.jp/signssm/. The pre-compiled binaries for some architectures are available in addition to the source code. The pre-installed binaries are also available on the Human Genome Center supercomputer system. The online manual and the supplementary information of SiGN-SSM is available on our web site. tamada@ims.u-tokyo.ac.jp.

KNBD: A Remote Kernel Block Server for Linux

Science.gov (United States)

Becker, Jeff

1999-01-01

I am developing a prototype of a Linux remote disk block server whose purpose is to serve as a lower level component of a parallel file system. Parallel file systems are an important component of high performance supercomputers and clusters. Although supercomputer vendors such as SGI and IBM have their own custom solutions, there has been a void and hence a demand for such a system on Beowulf-type PC Clusters. Recently, the Parallel Virtual File System (PVFS) project at Clemson University has begun to address this need (1). Although their system provides much of the functionality of (and indeed was inspired by) the equivalent file systems in the commercial supercomputer market, their system is all in user-space. Migrating their 10 services to the kernel could provide a performance boost, by obviating the need for expensive system calls. Thanks to Pavel Machek, the Linux kernel has provided the network block device (2) with kernels 2.1.101 and later. You can configure this block device to redirect reads and writes to a remote machine's disk. This can be used as a building block for constructing a striped file system across several nodes.

Summary of multi-core hardware and programming model investigations

Energy Technology Data Exchange (ETDEWEB)

Kelly, Suzanne Marie; Pedretti, Kevin Thomas Tauke; Levenhagen, Michael J.

2008-05-01

This report summarizes our investigations into multi-core processors and programming models for parallel scientific applications. The motivation for this study was to better understand the landscape of multi-core hardware, future trends, and the implications on system software for capability supercomputers. The results of this study are being used as input into the design of a new open-source light-weight kernel operating system being targeted at future capability supercomputers made up of multi-core processors. A goal of this effort is to create an agile system that is able to adapt to and efficiently support whatever multi-core hardware and programming models gain acceptance by the community.

Scientific visualization and radiology

International Nuclear Information System (INIS)

Lawrance, D.P.; Hoyer, C.E.; Wrestler, F.A.; Kuhn, M.J.; Moore, W.D.; Anderson, D.R.

1989-01-01

Scientific visualization is the visual presentation of numerical data. The National Center for Supercomputing Applications (NCSA) has developed methods for visualizing computerbased simulations of digital imaging data. The applicability of these various tools for unique and potentially medical beneficial display of MR images is investigated. Raw data are obtained from MR images of the brain, neck, spine, and brachial plexus obtained on a 1.5-T imager with multiple pulse sequences. A supercomputer and other mainframe resources run a variety of graphic and imaging programs using this data. An interdisciplinary team of imaging scientists, computer graphic programmers, an physicians works together to achieve useful information

How General-Purpose can a GPU be?

Directory of Open Access Journals (Sweden)

Philip Machanick

2015-12-01

Full Text Available The use of graphics processing units (GPUs in general-purpose computation (GPGPU is a growing field. GPU instruction sets, while implementing a graphics pipeline, draw from a range of single instruction multiple datastream (SIMD architectures characteristic of the heyday of supercomputers. Yet only one of these SIMD instruction sets has been of application on a wide enough range of problems to survive the era when the full range of supercomputer design variants was being explored: vector instructions. This paper proposes a reconceptualization of the GPU as a multicore design with minimal exotic modes of parallelism so as to make GPGPU truly general.

NIC symposium 2012. 25 years HLRZ/NIC. Proceedings

International Nuclear Information System (INIS)

Binder, Kurt

2012-01-01

Since 25 years the John von Neumann Institute for Computing (NIC), the former ''Hoechstleistungsrechenzentrum'', plays a pioneering role in supporting research in computational science at the fore-front, by giving large grants of computer time to carefully selected research projects. The scope of these projects ranges from fundamental aspects of physics, such as the physics of elementary particles and nuclear physics, astrophysics, statistical physics and physics of condensed matter, computational chemistry and life sciences, to more applied areas of research, such as the modelling of processes in the atmosphere, materials science, fluid dynamics applications in engineering, etc. Use of the supercomputer resources that the Juelich Supercomputing Centre (JSC) provides for these research projects. The present book, which appears in the framework of the biannual NIC Symposia series, continues a tradition started 10 years ago, to present selected highlights of this research to a broader audience. Due to space restrictions, only a small number of the research projects that are carried out at the NIC can be presented in this way. Projects that stand out as particularly excellent are nominated as ''John von Neumann Excellence Project'' by the review board. In 2010 this award was given to A. Muramatsu (Stuttgart) for his project on ''Quantum Monte Carlo studies of strongly correlated systems''. In 2011, two such awards were given to C. Hoelbling (Wuppertal) for his project ''Computing B K with 2+1 flavours at the physical mass point'', and another one to W. Paul (Halle) for ''Long range correlations at polymer-solid interfaces''. The procedures adopted by the NIC to identify the scientifically best projects for the allocation of computer time are of the same character as those used by organisations founded more recently, such as (in Germany) the Gauss Centre for Supercomputing (GCS), an alliance of the three German national supercomputing centres in Juelich, Garching and

NIC symposium 2012. 25 years HLRZ/NIC. Proceedings

Energy Technology Data Exchange (ETDEWEB)

Binder, Kurt [Mainz Univ. (Germany). Inst. fuer Physik; Muenster, Gernot [Muenster Univ. (Germany). Inst. fuer Theoretische Physik 1; Kremer, Manfred [Forschungszentrum Juelich GmbH (DE). Juelich Supercomputing Centre (JSC)

2012-08-07

Since 25 years the John von Neumann Institute for Computing (NIC), the former ''Hoechstleistungsrechenzentrum'', plays a pioneering role in supporting research in computational science at the fore-front, by giving large grants of computer time to carefully selected research projects. The scope of these projects ranges from fundamental aspects of physics, such as the physics of elementary particles and nuclear physics, astrophysics, statistical physics and physics of condensed matter, computational chemistry and life sciences, to more applied areas of research, such as the modelling of processes in the atmosphere, materials science, fluid dynamics applications in engineering, etc. Use of the supercomputer resources that the Juelich Supercomputing Centre (JSC) provides for these research projects. The present book, which appears in the framework of the biannual NIC Symposia series, continues a tradition started 10 years ago, to present selected highlights of this research to a broader audience. Due to space restrictions, only a small number of the research projects that are carried out at the NIC can be presented in this way. Projects that stand out as particularly excellent are nominated as ''John von Neumann Excellence Project'' by the review board. In 2010 this award was given to A. Muramatsu (Stuttgart) for his project on ''Quantum Monte Carlo studies of strongly correlated systems''. In 2011, two such awards were given to C. Hoelbling (Wuppertal) for his project ''Computing B{sub K} with 2+1 flavours at the physical mass point'', and another one to W. Paul (Halle) for ''Long range correlations at polymer-solid interfaces''. The procedures adopted by the NIC to identify the scientifically best projects for the allocation of computer time are of the same character as those used by organisations founded more recently, such as (in Germany) the Gauss Centre for Supercomputing (GCS), an alliance of the three German national supercomputing centres in Juelich, Garching

NIC symposium 2012. 25 years HLRZ/NIC. Proceedings

Energy Technology Data Exchange (ETDEWEB)

Binder, Kurt [Mainz Univ. (Germany). Inst. fuer Physik; Muenster, Gernot [Muenster Univ. (Germany). Inst. fuer Theoretische Physik 1; Kremer, Manfred (eds.) [Forschungszentrum Juelich GmbH (DE). Juelich Supercomputing Centre (JSC)

2012-08-07

Since 25 years the John von Neumann Institute for Computing (NIC), the former ''Hoechstleistungsrechenzentrum'', plays a pioneering role in supporting research in computational science at the fore-front, by giving large grants of computer time to carefully selected research projects. The scope of these projects ranges from fundamental aspects of physics, such as the physics of elementary particles and nuclear physics, astrophysics, statistical physics and physics of condensed matter, computational chemistry and life sciences, to more applied areas of research, such as the modelling of processes in the atmosphere, materials science, fluid dynamics applications in engineering, etc. Use of the supercomputer resources that the Juelich Supercomputing Centre (JSC) provides for these research projects. The present book, which appears in the framework of the biannual NIC Symposia series, continues a tradition started 10 years ago, to present selected highlights of this research to a broader audience. Due to space restrictions, only a small number of the research projects that are carried out at the NIC can be presented in this way. Projects that stand out as particularly excellent are nominated as ''John von Neumann Excellence Project'' by the review board. In 2010 this award was given to A. Muramatsu (Stuttgart) for his project on ''Quantum Monte Carlo studies of strongly correlated systems''. In 2011, two such awards were given to C. Hoelbling (Wuppertal) for his project ''Computing B{sub K} with 2+1 flavours at the physical mass point'', and another one to W. Paul (Halle) for ''Long range correlations at polymer-solid interfaces''. The procedures adopted by the NIC to identify the scientifically best projects for the allocation of computer time are of the same character as those used by organisations founded more recently, such as (in Germany) the Gauss Centre for

2016 ALCF Science Highlights

Energy Technology Data Exchange (ETDEWEB)

Collins, James R. [Argonne National Lab. (ANL), Argonne, IL (United States); Cerny, Beth A. [Argonne National Lab. (ANL), Argonne, IL (United States); Wolf, Laura [Argonne National Lab. (ANL), Argonne, IL (United States); Coffey, Richard M. [Argonne National Lab. (ANL), Argonne, IL (United States); Papka, Michael E. [Argonne National Lab. (ANL), Argonne, IL (United States)

2016-01-01

The Argonne Leadership Computing Facility provides supercomputing capabilities to the scientific and engineering community to advance fundamental discovery and understanding in a broad range of disciplines.

2015 Annual Report - Argonne Leadership Computing Facility

Energy Technology Data Exchange (ETDEWEB)

Collins, James R. [Argonne National Lab. (ANL), Argonne, IL (United States); Papka, Michael E. [Argonne National Lab. (ANL), Argonne, IL (United States); Cerny, Beth A. [Argonne National Lab. (ANL), Argonne, IL (United States); Coffey, Richard M. [Argonne National Lab. (ANL), Argonne, IL (United States)

2015-01-01

The Argonne Leadership Computing Facility provides supercomputing capabilities to the scientific and engineering community to advance fundamental discovery and understanding in a broad range of disciplines.

2014 Annual Report - Argonne Leadership Computing Facility

Energy Technology Data Exchange (ETDEWEB)

Collins, James R. [Argonne National Lab. (ANL), Argonne, IL (United States); Papka, Michael E. [Argonne National Lab. (ANL), Argonne, IL (United States); Cerny, Beth A. [Argonne National Lab. (ANL), Argonne, IL (United States); Coffey, Richard M. [Argonne National Lab. (ANL), Argonne, IL (United States)

2014-01-01

The Argonne Leadership Computing Facility provides supercomputing capabilities to the scientific and engineering community to advance fundamental discovery and understanding in a broad range of disciplines.

Ab initio molecular dynamics simulations for the role of hydrogen in catalytic reactions of furfural on Pd(111)

Science.gov (United States)

Xue, Wenhua; Dang, Hongli; Liu, Yingdi; Jentoft, Friederike; Resasco, Daniel; Wang, Sanwu

2014-03-01

In the study of catalytic reactions of biomass, furfural conversion over metal catalysts with the presence of hydrogen has attracted wide attention. We report ab initio molecular dynamics simulations for furfural and hydrogen on the Pd(111) surface at finite temperatures. The simulations demonstrate that the presence of hydrogen is important in promoting furfural conversion. In particular, hydrogen molecules dissociate rapidly on the Pd(111) surface. As a result of such dissociation, atomic hydrogen participates in the reactions with furfural. The simulations also provide detailed information about the possible reactions of hydrogen with furfural. Supported by DOE (DE-SC0004600). This research used the supercomputer resources of the XSEDE, the NERSC Center, and the Tandy Supercomputing Center.

Scalability of DL_POLY on High Performance Computing Platform

Directory of Open Access Journals (Sweden)

Mabule Samuel Mabakane

2017-12-01

Full Text Available This paper presents a case study on the scalability of several versions of the molecular dynamics code (DL_POLY performed on South Africa‘s Centre for High Performance Computing e1350 IBM Linux cluster, Sun system and Lengau supercomputers. Within this study different problem sizes were designed and the same chosen systems were employed in order to test the performance of DL_POLY using weak and strong scalability. It was found that the speed-up results for the small systems were better than large systems on both Ethernet and Infiniband network. However, simulations of large systems in DL_POLY performed well using Infiniband network on Lengau cluster as compared to e1350 and Sun supercomputer.

The Cure for HPC Neurosis: Multiple, Virtual Personalities!

Energy Technology Data Exchange (ETDEWEB)

Farber, Rob

2007-06-30

The selection of a new supercomputer for a scientific data center represents an interesting neurotic condition stemming from the conflict between a compulsion to acquire the best of the latest generation computer hardware, and unresolved issues as users seek validation from legacy scientific software - sometimes euphemistically called “research quality code”. Virtualization technology, now a mainstream feature on modern processors, permits multiple operating systems to efficiently and simultaneously run on each node of a supercomputer (or even your laptop and workstation). The benefits of this technology are many, ranging from supporting legacy software to paving the way towards robust petascale (1015 floating point operations per second) and eventually exascale (1018 floating point operations per second) computing.

1984 CERN school of computing

International Nuclear Information System (INIS)

1985-01-01

The eighth CERN School of Computing covered subjects mainly related to computing for elementary-particle physics. These proceedings contain written versions of most of the lectures delivered at the School. Notes on the following topics are included: trigger and data-acquisition plans for the LEP experiments; unfolding methods in high-energy physics experiments; Monte Carlo techniques; relational data bases; data networks and open systems; the Newcastle connection; portable operating systems; expert systems; microprocessors - from basic chips to complete systems; algorithms for parallel computers; trends in supercomputers and computational physics; supercomputing and related national projects in Japan; application of VLSI in high-energy physics, and single-user systems. See hints under the relevant topics. (orig./HSI)

Bringing ATLAS production to HPC resources. A case study with SuperMuc and Hydra

Energy Technology Data Exchange (ETDEWEB)

Duckeck, Guenter; Walker, Rodney [LMU Muenchen (Germany); Kennedy, John; Mazzaferro, Luca [RZG Garching (Germany); Kluth, Stefan [Max-Planck-Institut fuer Physik, Muenchen (Germany); Collaboration: ATLAS-Collaboration

2015-07-01

The possible usage of Supercomputer systems or HPC resources by ATLAS is now becoming viable due to the changing nature of these systems and it is also very attractive due to the need for increasing amounts of simulated data. The ATLAS experiment at CERN will begin a period of high luminosity data taking in 2015. The corresponding need for simulated data might potentially exceed the capabilities of the current Grid infrastructure. ATLAS aims to address this need by opportunistically accessing resources such as cloud and HPC systems. This contribution presents the results of two projects undertaken by LMU/LRZ and MPP/RZG to use the supercomputer facilities SuperMuc (LRZ) and Hydra (RZG). Both are Linux based supercomputers in the 100 k CPU-core category. The integration of such HPC resources into the ATLAS production system poses many challenges. Firstly, established techniques and features of standard WLCG operation are prohibited or much restricted on HPC systems, e.g. Grid middleware, software installation, outside connectivity, etc. Secondly, efficient use of available resources requires massive multi-core jobs, back-fill submission and check-pointing. We discuss the customization of these components and the strategies for HPC usage as well as possibilities for future directions.

NETL Super Computer

Data.gov (United States)

Federal Laboratory Consortium — The NETL Super Computer was designed for performing engineering calculations that apply to fossil energy research. It is one of the world’s larger supercomputers,...

Sadhana | Indian Academy of Sciences

Indian Academy of Sciences (India)

... VLSI clock interconnects; delay variability; PDF; process variation; Gaussian random ... Supercomputer Education and Research Centre, Indian Institute of Science, ... Manuscript received: 27 February 2009; Manuscript revised: 9 February ...

A Scheduling-Based Framework for Efficient Massively Parallel Execution, Phase I

Data.gov (United States)

National Aeronautics and Space Administration — The barrier to entry creating efficient, scalable applications for heterogeneous supercomputing environments is too high. EM Photonics has found that the majority of...

Oak Ridge Leadership Computing Facility (OLCF)

Data.gov (United States)

Federal Laboratory Consortium — The Oak Ridge Leadership Computing Facility (OLCF) was established at Oak Ridge National Laboratory in 2004 with the mission of standing up a supercomputer 100 times...

Metabolomics Workbench (MetWB)

Data.gov (United States)

U.S. Department of Health & Human Services — The Metabolomics Program's Data Repository and Coordinating Center (DRCC), housed at the San Diego Supercomputer Center (SDSC), University of California, San Diego,...

First-principles quantum-mechanical investigations: The role of water in catalytic conversion of furfural on Pd(111)

Science.gov (United States)

Xue, Wenhua; Borja, Miguel Gonzalez; Resasco, Daniel E.; Wang, Sanwu

2015-03-01

In the study of catalytic reactions of biomass, furfural conversion over metal catalysts with the presence of water has attracted wide attention. Recent experiments showed that the proportion of alcohol product from catalytic reactions of furfural conversion with palladium in the presence of water is significantly increased, when compared with other solvent including dioxane, decalin, and ethanol. We investigated the microscopic mechanism of the reactions based on first-principles quantum-mechanical calculations. We particularly identified the important role of water and the liquid/solid interface in furfural conversion. Our results provide atomic-scale details for the catalytic reactions. Supported by DOE (DE-SC0004600). This research used the supercomputer resources at NERSC, of XSEDE, at TACC, and at the Tandy Supercomputing Center.

[High energy particle physics]: Progress report covering the five year period from August 1, 1984 to May 31, 1989 with special emphasis for the period of August 1, 1988 to May 31, 1989: Part 1

International Nuclear Information System (INIS)

1989-01-01

In this document the High Energy Physics group reviews its accomplishments and progress during the past five years, with special emphasis for the past year and presents plans for continuing research during the next several years. During the last few years the effort of the experimental group has been divided approximately equally between fixed target physics and preparations for future collider experiments. The main emphasis of the theory group has been in the area of strong and electroweak phenomenology with an emphasis on hard scattering processes. With the recent creation of the Supercomputer Computations Research Institute, some work has also been done in the area of numerical simulations of condensed matter spin models and techniques for implementing numerical simulations on supercomputers

DCA++: A case for science driven application development for leadership computing platforms

International Nuclear Information System (INIS)

Summers, Michael S; Alvarez, Gonzalo; Meredith, Jeremy; Maier, Thomas A; Schulthess, Thomas C

2009-01-01

The DCA++ code was one of the early science applications that ran on jaguar at the National Center for Computational Sciences, and the first application code to sustain a petaflop/s under production conditions on a general-purpose supercomputer. The code implements a quantum cluster method with a Quantum Monte Carlo kernel to solve the 2D Hubbard model for high-temperature superconductivity. It is implemented in C++, making heavy use of the generic programming model. In this paper, we discuss how this code was developed, reaching scalability and high efficiency on the world's fastest supercomputer in only a few years. We show how the use of generic concepts combined with systematic refactoring of codes is a better strategy for computational sciences than a comprehensive upfront design.

The TESS Science Processing Operations Center

Science.gov (United States)

Jenkins, Jon M.; Twicken, Joseph D.; McCauliff, Sean; Campbell, Jennifer; Sanderfer, Dwight; Lung, David; Mansouri-Samani, Masoud; Girouard, Forrest; Tenenbaum, Peter; Klaus, Todd;

Numerical simulation of breakup and detachment of an axially ...

Indian Academy of Sciences (India)

Kishore Singh Patel

printing and lubrication for industrial application, colouring of fibres, wall coating for ...... Technology (DSTO 1329) and Supercomputing Education and Research ... mented technique for 3d moving boundary computations. J. Comput. Phys.

Los Alamos Science, Fall 1983 No. 9

Energy Technology Data Exchange (ETDEWEB)

Cooper, N G [ed.

1983-10-01

Topics covered in this issue include: cellular automata, gene expression, gen-bank and its promise for molecular genetics, and frontiers of supercomputing. Abstracts have been prepared for the individual items. (GHT)

Parallel Earthquake Simulations on Large-Scale Multicore Supercomputers

KAUST Repository

Wu, Xingfu

2011-01-01

Earthquakes are one of the most destructive natural hazards on our planet Earth. Hugh earthquakes striking offshore may cause devastating tsunamis, as evidenced by the 11 March 2011 Japan (moment magnitude Mw9.0) and the 26 December 2004 Sumatra (Mw9.1) earthquakes. Earthquake prediction (in terms of the precise time, place, and magnitude of a coming earthquake) is arguably unfeasible in the foreseeable future. To mitigate seismic hazards from future earthquakes in earthquake-prone areas, such as California and Japan, scientists have been using numerical simulations to study earthquake rupture propagation along faults and seismic wave propagation in the surrounding media on ever-advancing modern computers over past several decades. In particular, ground motion simulations for past and future (possible) significant earthquakes have been performed to understand factors that affect ground shaking in populated areas, and to provide ground shaking characteristics and synthetic seismograms for emergency preparation and design of earthquake-resistant structures. These simulation results can guide the development of more rational seismic provisions for leading to safer, more efficient, and economical50pt]Please provide V. Taylor author e-mail ID. structures in earthquake-prone regions.

Parallel Earthquake Simulations on Large-Scale Multicore Supercomputers

KAUST Repository

Wu, Xingfu; Duan, Benchun; Taylor, Valerie

2011-01-01

, such as California and Japan, scientists have been using numerical simulations to study earthquake rupture propagation along faults and seismic wave propagation in the surrounding media on ever-advancing modern computers over past several decades. In particular

Radiation-magnetohydrodynamics of fusion plasmas on parallel supercomputers

International Nuclear Information System (INIS)

Yasar, O.; Moses, G.A.; Tautges, T.J.

1993-01-01

A parallel computational model to simulate fusion plasmas in the radiation-magnetohydrodynamics (R-MHD) framework is presented. Plasmas are often treated in a fluid dynamics context (magnetohydrodynamics, MHD), but when the flow field is coupled with the radiation field it falls into a more complex category, radiation magnetohydrodynamics (R-MHD), where the interaction between the flow field and the radiation field is nonlinear. The solution for the radiation field usually dominates the R-MHD computation. To solve for the radiation field, one usually chooses the S N discrete ordinates method (a deterministic method) rather than the Monte Carlo method if the geometry is not complex. The discrete ordinates method on a massively parallel processor (Intel iPSC/860) is implemented. The speedup is 14 for a run on 16 processors and the performance is 3.7 times better than a single CRAY YMP processor implementation. (orig./DG)

Application of supercomputers to 3-D mantle convection

International Nuclear Information System (INIS)

Baumgardner, J.R.

1986-01-01

Current generation vector machines are providing for the first time the computing power needed to treat planetary mantle convection in a fully three-dimensional fashion. A numerical technique known as multigrid has been implemented in spherical geometry using a hierarchy of meshes constructed from the regular icosahedron to yield a highly efficient three-dimensional compressible Eulerian finite element hydrodynamics formulation. The paper describes the numerical method and presents convection solutions for the mantles of both the earth and the Moon. In the case of the Earth, the convection pattern is characterized by upwelling in narrow circular plumes originating at the core-mantle boundary and by downwelling in sheets or slabs derived from the cold upper boundary layer. The preferred number of plumes appears to be on the order of six or seven. For the Moon, the numerical results indicate that development of a predominately L = 2 pattern in later lunar history is a plausible explanation for the present large second-degree non-hydrostatic component in the lunar figure

ATLAS FTK a - very complex - custom parallel supercomputer

CERN Document Server

Kimura, Naoki; The ATLAS collaboration

2016-01-01

In the ever increasing pile-up LHC environment advanced techniques of analysing the data are implemented in order to increase the rate of relevant physics processes with respect to background processes. The Fast TracKer (FTK) is a track finding implementation at hardware level that is designed to deliver full-scan tracks with $p_{T}$ above 1GeV to the ATLAS trigger system for every L1 accept (at a maximum rate of 100kHz). In order to achieve this performance a highly parallel system was designed and now it is under installation in ATLAS. In the beginning of 2016 it will provide tracks for the trigger system in a region covering the central part of the ATLAS detector, and during the year it's coverage will be extended to the full detector coverage. The system relies on matching hits coming from the silicon tracking detectors against 1 billion patterns stored in specially designed ASICS chips (Associative memory - AM06). In a first stage coarse resolution hits are matched against the patterns and the accepted h...

Accelerating Communication-Intensive Applications via Novel Data Compression Techniques, Phase I

Data.gov (United States)

National Aeronautics and Space Administration — Processor speed has traditionally grown at a rate faster than that of communication speed in computer and supercomputer networks, and it is expected that this trend...

Sadhana | Indian Academy of Sciences

Indian Academy of Sciences (India)

Checkpointing is the process of saving the status information. ... Supercomputer Education and Research Centre (SERC), Indian Institute of Science, Bangalore 560 ... Manuscript received: 27 August 1998; Manuscript revised: 8 June 2000 ...

Web interface for plasma analysis codes

Energy Technology Data Exchange (ETDEWEB)

Emoto, M. [National Institute for Fusion Science, 322-6 Oroshi, Toki, Gifu 509-5292 (Japan)], E-mail: emo@nifs.ac.jp; Murakami, S. [Kyoto University, Yoshida-Honmachi, Sakyo-ku, Kyoto 606-8501 (Japan); Yoshida, M.; Funaba, H.; Nagayama, Y. [National Institute for Fusion Science, 322-6 Oroshi, Toki, Gifu 509-5292 (Japan)

2008-04-15

There are many analysis codes that analyze various aspects of plasma physics. However, most of them are FORTRAN programs that are written to be run in supercomputers. On the other hand, many scientists use GUI (graphical user interface)-based operating systems. For those who are not familiar with supercomputers, it is a difficult task to run analysis codes in supercomputers, and they often hesitate to use these programs to substantiate their ideas. Furthermore, these analysis codes are written for personal use, and the programmers do not expect these programs to be run by other users. In order to make these programs to be widely used by many users, the authors developed user-friendly interfaces using a Web interface. Since the Web browser is one of the most common applications, it is useful for both the users and developers. In order to realize interactive Web interface, AJAX technique is widely used, and the authors also adopted AJAX. To build such an AJAX based Web system, Ruby on Rails plays an important role in this system. Since this application framework, which is written in Ruby, abstracts the Web interfaces necessary to implement AJAX and database functions, it enables the programmers to efficiently develop the Web-based application. In this paper, the authors will introduce the system and demonstrate the usefulness of this approach.

NIC symposium 2010. Proceedings

Energy Technology Data Exchange (ETDEWEB)

Muenster, Gernot [Muenster Univ. (Germany). Inst. fuer Theoretische Physik 1; Wolf, Dietrich [Duisburg-Essen Univ., Duisburg (Germany). Fakultaet fuer Physik; Kremer, Manfred (eds.) [Forschungszentrum Juelich GmbH (DE). Juelich Supercomputing Centre (JSC)

2012-06-21

The fifth NIC-Symposium gave an overview of the activities of the John von Neumann Institute for Computing (NIC) and of the results obtained in the last two years by research groups supported by the NIC. The large recent progress in supercomputing is highlighted by the fact that the newly installed Blue Gene/P system in Juelich - with a peak performance of 1 Petaflop/s - currently ranks number four in the TOP500 list. This development opens new dimensions in simulation science for researchers in Germany and Europe. NIC - a joint foundation of Forschungszentrum Juelich, Deutsches Elektronen-Synchrotron (DESY) and Gesellschaft fuer Schwerionenforschung (GSI) - supports with its members' supercomputer facilities about 130 research groups at universities and national labs working on computer simulations in various fields of science. Fifteen invited lectures covered selected topics in the following fields: Astrophysics Biophysics Chemistry Elementary Particle Physics Condensed Matter Materials Science Soft Matter Science Environmental Research Hydrodynamics and turbulence Plasma Physics Computer Science The talks are intended to inform a broad audience of scientists and the interested public about the research activities at NIC. The proceedings of the symposium cover projects that have been supported by the IBM supercomputers JUMP and IBM Blue Gene/P in Juelich and the APE topical computer at DESY-Zeuthen in an even wider range than the lectures.

NIC symposium 2010. Proceedings

Energy Technology Data Exchange (ETDEWEB)

Muenster, Gernot [Muenster Univ. (Germany). Inst. fuer Theoretische Physik 1; Wolf, Dietrich [Duisburg-Essen Univ., Duisburg (Germany). Fakultaet fuer Physik; Kremer, Manfred [Forschungszentrum Juelich GmbH (DE). Juelich Supercomputing Centre (JSC)

2012-06-21

The fifth NIC-Symposium gave an overview of the activities of the John von Neumann Institute for Computing (NIC) and of the results obtained in the last two years by research groups supported by the NIC. The large recent progress in supercomputing is highlighted by the fact that the newly installed Blue Gene/P system in Juelich - with a peak performance of 1 Petaflop/s - currently ranks number four in the TOP500 list. This development opens new dimensions in simulation science for researchers in Germany and Europe. NIC - a joint foundation of Forschungszentrum Juelich, Deutsches Elektronen-Synchrotron (DESY) and Gesellschaft fuer Schwerionenforschung (GSI) - supports with its members' supercomputer facilities about 130 research groups at universities and national labs working on computer simulations in various fields of science. Fifteen invited lectures covered selected topics in the following fields: Astrophysics Biophysics Chemistry Elementary Particle Physics Condensed Matter Materials Science Soft Matter Science Environmental Research Hydrodynamics and turbulence Plasma Physics Computer Science The talks are intended to inform a broad audience of scientists and the interested public about the research activities at NIC. The proceedings of the symposium cover projects that have been supported by the IBM supercomputers JUMP and IBM Blue Gene/P in Juelich and the APE topical computer at DESY-Zeuthen in an even wider range than the lectures.

NIC symposium 2010. Proceedings

International Nuclear Information System (INIS)

Muenster, Gernot

2012-01-01

The fifth NIC-Symposium gave an overview of the activities of the John von Neumann Institute for Computing (NIC) and of the results obtained in the last two years by research groups supported by the NIC. The large recent progress in supercomputing is highlighted by the fact that the newly installed Blue Gene/P system in Juelich - with a peak performance of 1 Petaflop/s - currently ranks number four in the TOP500 list. This development opens new dimensions in simulation science for researchers in Germany and Europe. NIC - a joint foundation of Forschungszentrum Juelich, Deutsches Elektronen-Synchrotron (DESY) and Gesellschaft fuer Schwerionenforschung (GSI) - supports with its members' supercomputer facilities about 130 research groups at universities and national labs working on computer simulations in various fields of science. Fifteen invited lectures covered selected topics in the following fields: Astrophysics Biophysics Chemistry Elementary Particle Physics Condensed Matter Materials Science Soft Matter Science Environmental Research Hydrodynamics and turbulence Plasma Physics Computer Science The talks are intended to inform a broad audience of scientists and the interested public about the research activities at NIC. The proceedings of the symposium cover projects that have been supported by the IBM supercomputers JUMP and IBM Blue Gene/P in Juelich and the APE topical computer at DESY-Zeuthen in an even wider range than the lectures.

Web interface for plasma analysis codes

International Nuclear Information System (INIS)

Emoto, M.; Murakami, S.; Yoshida, M.; Funaba, H.; Nagayama, Y.

2008-01-01

There are many analysis codes that analyze various aspects of plasma physics. However, most of them are FORTRAN programs that are written to be run in supercomputers. On the other hand, many scientists use GUI (graphical user interface)-based operating systems. For those who are not familiar with supercomputers, it is a difficult task to run analysis codes in supercomputers, and they often hesitate to use these programs to substantiate their ideas. Furthermore, these analysis codes are written for personal use, and the programmers do not expect these programs to be run by other users. In order to make these programs to be widely used by many users, the authors developed user-friendly interfaces using a Web interface. Since the Web browser is one of the most common applications, it is useful for both the users and developers. In order to realize interactive Web interface, AJAX technique is widely used, and the authors also adopted AJAX. To build such an AJAX based Web system, Ruby on Rails plays an important role in this system. Since this application framework, which is written in Ruby, abstracts the Web interfaces necessary to implement AJAX and database functions, it enables the programmers to efficiently develop the Web-based application. In this paper, the authors will introduce the system and demonstrate the usefulness of this approach

Validation of the GPU-Accelerated CFD Solver ELBE for Free Surface Flow Problems in Civil and Environmental Engineering

Directory of Open Access Journals (Sweden)

Christian F. Janßen

2015-07-01

Full Text Available This contribution is dedicated to demonstrating the high potential and manifold applications of state-of-the-art computational fluid dynamics (CFD tools for free-surface flows in civil and environmental engineering. All simulations were performed with the academic research code ELBE (efficient lattice boltzmann environment, http://www.tuhh.de/elbe. The ELBE code follows the supercomputing-on-the-desktop paradigm and is especially designed for local supercomputing, without tedious accesses to supercomputers. ELBE uses graphics processing units (GPU to accelerate the computations and can be used in a single GPU-equipped workstation of, e.g., a design engineer. The code has been successfully validated in very different fields, mostly related to naval architecture and mechanical engineering. In this contribution, we give an overview of past and present applications with practical relevance for civil engineers. The presented applications are grouped into three major categories: (i tsunami simulations, considering wave propagation, wave runup, inundation and debris flows; (ii dam break simulations; and (iii numerical wave tanks for the calculation of hydrodynamic loads on fixed and moving bodies. This broad range of applications in combination with accurate numerical results and very competitive times to solution demonstrates that modern CFD tools in general, and the ELBE code in particular, can be a helpful design tool for civil and environmental engineers.

Molecular dynamics beyonds the limits: Massive scaling on 72 racks of a BlueGene/P and supercooled glass dynamics of a 1 billion particles system

KAUST Repository

Allsopp, Nicholas; Ruocco, Giancarlo; Fratalocchi, Andrea

2012-01-01

We report scaling results on the world's largest supercomputer of our recently developed Billions-Body Molecular Dynamics (BBMD) package, which was especially designed for massively parallel simulations of the short-range atomic dynamics

Federal High End Computing (HEC) Information Portal

Data.gov (United States)

Networking and Information Technology Research and Development, Executive Office of the President — This portal provides information about opportunities to engage in U.S. Federal government high performance computing activities, including supercomputer use,...

Why genetic information processing could have a quantum basis

Indian Academy of Sciences (India)

Unknown

Centre for Theoretical Studies and Supercomputer Education and Research Centre, ... the parent to the offspring, sensory information conveyed by the sense organ to the .... The task involved in genetic information processing is. ASSEMBLY.

Avoiding acidic region streaking in two-dimensional gel ...

Indian Academy of Sciences (India)

2014-07-21

Jul 21, 2014 ... 1Bioinformatics Centre, Supercomputer Education and Research Centre and 2Department of ... has made 2DE a popular instrument in modern proteomics ...... Proteomics study on the hepatoprotective effects of traditional.

Grid Computing

Indian Academy of Sciences (India)

IAS Admin

emergence of supercomputers led to the use of computer simula- tion as an .... Scientific and engineering applications (e.g., Tera grid secure gate way). Collaborative ... Encryption, privacy, protection from malicious software. Physical Layer.

Scalable I/O Systems via Node-Local Storage: Approaching 1 TB/sec File I/O

Energy Technology Data Exchange (ETDEWEB)

Bronevetsky, G; Moody, A

2009-08-18

In the race to PetaFLOP-speed supercomputing systems, the increase in computational capability has been accompanied by corresponding increases in CPU count, total RAM, and storage capacity. However, a proportional increase in storage bandwidth has lagged behind. In order to improve system reliability and to reduce maintenance effort for modern large-scale systems, system designers have opted to remove node-local storage from the compute nodes. Today's multi-TeraFLOP supercomputers are typically attached to parallel file systems that provide only tens of GBs/s of I/O bandwidth. As a result, such machines have access to much less than 1GB/s of I/O bandwidth per TeraFLOP of compute power, which is below the generally accepted limit required for a well-balanced system. In a many ways, the current I/O bottleneck limits the capabilities of modern supercomputers, specifically in terms of limiting their working sets and restricting fault tolerance techniques, which become critical on systems consisting of tens of thousands of components. This paper resolves the dilemma between high performance and high reliability by presenting an alternative system design which makes use of node-local storage to improve aggregate system I/O bandwidth. In this work, we focus on the checkpointing use-case and present an experimental evaluation of the Scalable Checkpoint/Restart (SCR) library, a new adaptive checkpointing library that uses node-local storage to significantly improve the checkpointing performance of large-scale supercomputers. Experiments show that SCR achieves unprecedented write speeds, reaching a measured 700GB/s of aggregate bandwidth on 8,752 processors and an estimated 1TB/s for a similarly structured machine of 12,500 processors. This corresponds to a speedup of over 70x compared to the bandwidth provided by the 10GB/s parallel file system the cluster uses. Further, SCR can adapt to an environment in which there is wide variation in performance or capacity among

Engineering-Based Thermal CFD Simulations on Massive Parallel Systems

KAUST Repository

Frisch, Jé rô me; Mundani, Ralf-Peter; Rank, Ernst; van Treeck, Christoph

2015-01-01

The development of parallel Computational Fluid Dynamics (CFD) codes is a challenging task that entails efficient parallelization concepts and strategies in order to achieve good scalability values when running those codes on modern supercomputers

Testing quantum dynamics in genetic information processing

Indian Academy of Sciences (India)

Unknown

Centre for Theoretical Studies, and Supercomputer Education and Research Centre,. Indian Institute .... values of a and comparing their replication rates, we can experimentally ... This is a substantial tolerance margin for the quantum algorithm ...

BigData and computing challenges in high energy and nuclear physics

Science.gov (United States)

Klimentov, A.; Grigorieva, M.; Kiryanov, A.; Zarochentsev, A.

2017-06-01

In this contribution we discuss the various aspects of the computing resource needs experiments in High Energy and Nuclear Physics, in particular at the Large Hadron Collider. This will evolve in the future when moving from LHC to HL-LHC in ten years from now, when the already exascale levels of data we are processing could increase by a further order of magnitude. The distributed computing environment has been a great success and the inclusion of new super-computing facilities, cloud computing and volunteering computing for the future is a big challenge, which we are successfully mastering with a considerable contribution from many super-computing centres around the world, academic and commercial cloud providers. We also discuss R&D computing projects started recently in National Research Center ``Kurchatov Institute''

DCA++: A case for science driven application development for leadership computing platforms

Energy Technology Data Exchange (ETDEWEB)

Summers, Michael S; Alvarez, Gonzalo; Meredith, Jeremy; Maier, Thomas A [Computer Science and Mathematics Division, Oak Ridge National Laboratory, P. O. Box 2008, Mail Stop 6164, Oak Ridge, TN 37831 (United States); Schulthess, Thomas C, E-mail: schulthess@cscs.c [Swiss National Supercomputer Center and Institute for Theoretical Physics, ETH Zurich, CSCS MAN E 133, Galeria 2, CH-9628 Manno (Switzerland)

2009-07-01

The DCA++ code was one of the early science applications that ran on jaguar at the National Center for Computational Sciences, and the first application code to sustain a petaflop/s under production conditions on a general-purpose supercomputer. The code implements a quantum cluster method with a Quantum Monte Carlo kernel to solve the 2D Hubbard model for high-temperature superconductivity. It is implemented in C++, making heavy use of the generic programming model. In this paper, we discuss how this code was developed, reaching scalability and high efficiency on the world's fastest supercomputer in only a few years. We show how the use of generic concepts combined with systematic refactoring of codes is a better strategy for computational sciences than a comprehensive upfront design.

C Versus Fortran-77 for Scientific Programming

Directory of Open Access Journals (Sweden)

Tom MacDonald

1992-01-01

Full Text Available The predominant programming language for numeric and scientific applications is Fortran-77 and supercomputers are primarily used to run large-scale numeric and scientific applications. Standard C* is not widely used for numerical and scientific programming, yet Standard C provides many desirable linguistic features not present in Fortran-77. Furthermore, the existence of a standard library and preprocessor eliminates the worst portability problems. A comparison of Standard C and Fortran-77 shows several key deficiencies in C that reduce its ability to adequately solve some numerical problems. Some of these problems have already been addressed by the C standard but others remain. Standard C with a few extensions and modifications could be suitable for all numerical applications and could become more popular in supercomputing environments.

Visualization of numerically simulated aerodynamic flow fields

International Nuclear Information System (INIS)

Hian, Q.L.; Damodaran, M.

1991-01-01

The focus of this paper is to describe the development and the application of an interactive integrated software to visualize numerically simulated aerodynamic flow fields so as to enable the practitioner of computational fluid dynamics to diagnose the numerical simulation and to elucidate essential flow physics from the simulation. The input to the software is the numerical database crunched by a supercomputer and typically consists of flow variables and computational grid geometry. This flow visualization system (FVS), written in C language is targetted at the Personal IRIS Workstations. In order to demonstrate the various visualization modules, the paper also describes the application of this software to visualize two- and three-dimensional flow fields past aerodynamic configurations which have been numerically simulated on the NEC-SXIA Supercomputer. 6 refs

BigData and computing challenges in high energy and nuclear physics

International Nuclear Information System (INIS)

Klimentov, A.; Grigorieva, M.; Kiryanov, A.; Zarochentsev, A.

2017-01-01

In this contribution we discuss the various aspects of the computing resource needs experiments in High Energy and Nuclear Physics, in particular at the Large Hadron Collider. This will evolve in the future when moving from LHC to HL-LHC in ten years from now, when the already exascale levels of data we are processing could increase by a further order of magnitude. The distributed computing environment has been a great success and the inclusion of new super-computing facilities, cloud computing and volunteering computing for the future is a big challenge, which we are successfully mastering with a considerable contribution from many super-computing centres around the world, academic and commercial cloud providers. We also discuss R and D computing projects started recently in National Research Center ''Kurchatov Institute''

Partial Overhaul and Initial Parallel Optimization of KINETICS, a Coupled Dynamics and Chemistry Atmosphere Model

Science.gov (United States)

Nguyen, Howard; Willacy, Karen; Allen, Mark

2012-01-01

KINETICS is a coupled dynamics and chemistry atmosphere model that is data intensive and computationally demanding. The potential performance gain from using a supercomputer motivates the adaptation from a serial version to a parallelized one. Although the initial parallelization had been done, bottlenecks caused by an abundance of communication calls between processors led to an unfavorable drop in performance. Before starting on the parallel optimization process, a partial overhaul was required because a large emphasis was placed on streamlining the code for user convenience and revising the program to accommodate the new supercomputers at Caltech and JPL. After the first round of optimizations, the partial runtime was reduced by a factor of 23; however, performance gains are dependent on the size of the data, the number of processors requested, and the computer used.

Tools for 3D scientific visualization in computational aerodynamics at NASA Ames Research Center

International Nuclear Information System (INIS)

Bancroft, G.; Plessel, T.; Merritt, F.; Watson, V.

1989-01-01

Hardware, software, and techniques used by the Fluid Dynamics Division (NASA) for performing visualization of computational aerodynamics, which can be applied to the visualization of flow fields from computer simulations of fluid dynamics about the Space Shuttle, are discussed. Three visualization techniques applied, post-processing, tracking, and steering, are described, as well as the post-processing software packages used, PLOT3D, SURF (Surface Modeller), GAS (Graphical Animation System), and FAST (Flow Analysis software Toolkit). Using post-processing methods a flow simulation was executed on a supercomputer and, after the simulation was complete, the results were processed for viewing. It is shown that the high-resolution, high-performance three-dimensional workstation combined with specially developed display and animation software provides a good tool for analyzing flow field solutions obtained from supercomputers. 7 refs

Physicists set new record for network data transfer

CERN Multimedia

2007-01-01

"An international team of physicists, computer scientists, and network engineers joined forces to set new records for sustained data transfer between storage systems durint the SuperComputing 2006 (SC06) Bandwidth Challenge (BWC). (3 pages)

Cytoview: Development of a cell modelling framework

Indian Academy of Sciences (India)

PRAKASH KUMAR

is an important aspect of cell modelling. ... 1Supercomputer Education and Research Centreand 2Bioinformatics Centre, Indian Institute ... Important aspects in each panel are listed. ... subsumption relationship, in which the child term is a more.

Effective Analysis of NGS Metagenomic Data with Ultra-Fast Clustering Algorithms (MICW - Metagenomics Informatics Challenges Workshop: 10K Genomes at a Time)

Energy Technology Data Exchange (ETDEWEB)

Li, Weizhong

2011-10-12

San Diego Supercomputer Center's Weizhong Li on "Effective Analysis of NGS Metagenomic Data with Ultra-fast Clustering Algorithms" at the Metagenomics Informatics Challenges Workshop held at the DOE JGI on October 12-13, 2011.

Gone in eight seconds: Canadian data-transfer record points to the future of the Internet

CERN Multimedia

Tam, P

2002-01-01

"When completed in 2007, a new grid network will harness the processing power of many machines across Canada to create a communal supercomputer. It will be tailor-made for researchers with high-performance computing needs" (1 page).

Quantum Transport Simulations of Nanoscale Materials

KAUST Repository

Obodo, Tobechukwu Joshua

2016-01-01

-performance supercomputers allow us to control and exploit their microscopic properties at the atomic scale, hence making it possible to design novel nanoscale molecular devices with interesting features (e.g switches, rectifiers, negative differential conductance, and high

Even bigger days are ahead for data superhighway

CERN Multimedia

Kotulak, R

2003-01-01

"Scientists at Fermi and Argonne National Laboratories, along with colleagues around the globe, are building the ultimate supercomputer that they envision almost anyone eventually could use to tackle tough problems now out of their reach" (1 page).

Numerical simulations of topological and correlated quantum matter

Energy Technology Data Exchange (ETDEWEB)

Assaad, Fakher F. [Wuerzburg Univ. (Germany). Inst. fuer Theoretische Physik und Astrophysik

2016-11-01

The complexity of the solid state does not allow us to carry out simulations of correlated materials without adopting approximation schemes. In this project we are tackling this daunting task with complementary techniques. On one hand one can start with density functional theory in the local density approximation and then add dynamical local interactions using the so called dynamical mean-field approximation. This approach has the merit of being material dependent in the sense that it is possible to include the specific chemical constituents of the material under investigation. Progress in this domain is described below. Another venue is to concentrate on phenomena occurring in a class of materials. Here, the strategy is to define models which one can simulate in polynomial time on supercomputing architectures, and which reproduce the phenomena under investigation. This route has been remarkably successful, and we are now in a position to provide controlled model calculations which can cope with antiferromagnetic fluctuations in metals, or nematic instabilities of fermi liquids. Both phenomena are crucial for our understanding of high temperature superconductivity in the cuprates and the pnictides. Access to the LRZ supercomputing center was imperative during the current grant period to do the relevant simulations on a wide range of topics on correlated electrons. In all cases access to supercomputing facilities allows to carry out simulations on larger and larger system sizes so as to be able to extrapolate to the thermodynamic limit relevant for the understanding of experiments and collective phenomena.

Computation Directorate 2008 Annual Report

Energy Technology Data Exchange (ETDEWEB)

Crawford, D L

2009-03-25

Whether a computer is simulating the aging and performance of a nuclear weapon, the folding of a protein, or the probability of rainfall over a particular mountain range, the necessary calculations can be enormous. Our computers help researchers answer these and other complex problems, and each new generation of system hardware and software widens the realm of possibilities. Building on Livermore's historical excellence and leadership in high-performance computing, Computation added more than 331 trillion floating-point operations per second (teraFLOPS) of power to LLNL's computer room floors in 2008. In addition, Livermore's next big supercomputer, Sequoia, advanced ever closer to its 2011-2012 delivery date, as architecture plans and the procurement contract were finalized. Hyperion, an advanced technology cluster test bed that teams Livermore with 10 industry leaders, made a big splash when it was announced during Michael Dell's keynote speech at the 2008 Supercomputing Conference. The Wall Street Journal touted Hyperion as a 'bright spot amid turmoil' in the computer industry. Computation continues to measure and improve the costs of operating LLNL's high-performance computing systems by moving hardware support in-house, by measuring causes of outages to apply resources asymmetrically, and by automating most of the account and access authorization and management processes. These improvements enable more dollars to go toward fielding the best supercomputers for science, while operating them at less cost and greater responsiveness to the customers.

Understanding Lustre Internals

Energy Technology Data Exchange (ETDEWEB)

Wang, Feiyi [ORNL; Oral, H Sarp [ORNL; Shipman, Galen M [ORNL; Drokin, Oleg [ORNL; Wang, Di [ORNL; Huang, He [ORNL

2009-04-01

Lustre was initiated and funded, almost a decade ago, by the U.S. Department of Energy (DoE) Office of Science and National Nuclear Security Administration laboratories to address the need for an open source, highly-scalable, high-performance parallel filesystem on by then present and future supercomputing platforms. Throughout the last decade, it was deployed over numerous medium-to-large-scale supercomputing platforms and clusters, and it performed and met the expectations of the Lustre user community. As it stands at the time of writing this document, according to the Top500 list, 15 of the top 30 supercomputers in the world use Lustre filesystem. This report aims to present a streamlined overview on how Lustre works internally at reasonable details including relevant data structures, APIs, protocols and algorithms involved for Lustre version 1.6 source code base. More importantly, it tries to explain how various components interconnect with each other and function as a system. Portions of this report are based on discussions with Oak Ridge National Laboratory Lustre Center of Excellence team members and portions of it are based on our own understanding of how the code works. We, as the authors team bare all responsibilities for all errors and omissions in this document. We can only hope it helps current and future Lustre users and Lustre code developers as much as it helped us understanding the Lustre source code and its internal workings.

grid will help physicists' global hunt for particles Researchers have begun running experiments with the MidWest Tier 2 Center, one of five regional computing centers in the US.

CERN Multimedia

Ames, Ben

2006-01-01

"When physicists at Switzerland's CERN laboratory turn on their newsest particle collider in 2007, they will rely on computer scientists in Chicago and Indianapolis to help sift through the results using a worldwide supercomputing grid." (1/2 page)

Real World Uses For Nagios APIs

Science.gov (United States)

Singh, Janice

2014-01-01

This presentation describes the Nagios 4 APIs and how the NASA Advanced Supercomputing at Ames Research Center is employing them to upgrade its graphical status display (the HUD) and explain why it's worth trying to use them yourselves.

Computer codes for designing proton linear accelerators

International Nuclear Information System (INIS)

Kato, Takao

1992-01-01

Computer codes for designing proton linear accelerators are discussed from the viewpoint of not only designing but also construction and operation of the linac. The codes are divided into three categories according to their purposes: 1) design code, 2) generation and simulation code, and 3) electric and magnetic fields calculation code. The role of each category is discussed on the basis of experience at KEK (the design of the 40-MeV proton linac and its construction and operation, and the design of the 1-GeV proton linac). We introduce our recent work relevant to three-dimensional calculation and supercomputer calculation: 1) tuning of MAFIA (three-dimensional electric and magnetic fields calculation code) for supercomputer, 2) examples of three-dimensional calculation of accelerating structures by MAFIA, 3) development of a beam transport code including space charge effects. (author)

Implementation, capabilities, and benchmarking of Shift, a massively parallel Monte Carlo radiation transport code

International Nuclear Information System (INIS)

Pandya, Tara M.; Johnson, Seth R.; Evans, Thomas M.; Davidson, Gregory G.; Hamilton, Steven P.; Godfrey, Andrew T.

2015-01-01

This paper discusses the implementation, capabilities, and validation of Shift, a massively parallel Monte Carlo radiation transport package developed and maintained at Oak Ridge National Laboratory. It has been developed to scale well from laptop to small computing clusters to advanced supercomputers. Special features of Shift include hybrid capabilities for variance reduction such as CADIS and FW-CADIS, and advanced parallel decomposition and tally methods optimized for scalability on supercomputing architectures. Shift has been validated and verified against various reactor physics benchmarks and compares well to other state-of-the-art Monte Carlo radiation transport codes such as MCNP5, CE KENO-VI, and OpenMC. Some specific benchmarks used for verification and validation include the CASL VERA criticality test suite and several Westinghouse AP1000 ® problems. These benchmark and scaling studies show promising results

The QCDOC Project

International Nuclear Information System (INIS)

Boyle, P.; Chen, D.; Christ, N.; Clark, M.; Cohen, S.; Cristian, C.; Dong, Z.; Gara, A.; Joo, B.; Jung, C.; Kim, C.; Levkova, L.; Liao, X.; Liu, G.; Li, S.; Lin, H.; Mawhinney, R.; Ohta, S.; Petrov, K.; Wettig, T.; Yamaguchi, A.

2005-01-01

The QCDOC project has developed a supercomputer optimised for the needs of Lattice QCD simulations. It provides a very competitive price to sustained performance ratio of around $1 USD per sustained Megaflop/s in combination with outstanding scalability. Thus very large systems delivering over 5 TFlop/s of performance on the evolution of a single lattice is possible. Large prototypes have been built and are functioning correctly. The software environment raises the state of the art in such custom supercomputers. It is based on a lean custom node operating system that eliminates many unnecessary overheads that plague other systems. Despite the custom nature, the operating system implements a standards compliant UNIX-like programming environment easing the porting of software from other systems. The SciDAC QMP interface adds internode communication in a fashion that provides a uniform cross-platform programming environment

Multi-GPU accelerated three-dimensional FDTD method for electromagnetic simulation.

Science.gov (United States)

Nagaoka, Tomoaki; Watanabe, Soichi

2011-01-01

Numerical simulation with a numerical human model using the finite-difference time domain (FDTD) method has recently been performed in a number of fields in biomedical engineering. To improve the method's calculation speed and realize large-scale computing with the numerical human model, we adapt three-dimensional FDTD code to a multi-GPU environment using Compute Unified Device Architecture (CUDA). In this study, we used NVIDIA Tesla C2070 as GPGPU boards. The performance of multi-GPU is evaluated in comparison with that of a single GPU and vector supercomputer. The calculation speed with four GPUs was approximately 3.5 times faster than with a single GPU, and was slightly (approx. 1.3 times) slower than with the supercomputer. Calculation speed of the three-dimensional FDTD method using GPUs can significantly improve with an expanding number of GPUs.

An efficient parallel algorithm for matrix-vector multiplication

Energy Technology Data Exchange (ETDEWEB)

Hendrickson, B.; Leland, R.; Plimpton, S.

1993-03-01

The multiplication of a vector by a matrix is the kernel computation of many algorithms in scientific computation. A fast parallel algorithm for this calculation is therefore necessary if one is to make full use of the new generation of parallel supercomputers. This paper presents a high performance, parallel matrix-vector multiplication algorithm that is particularly well suited to hypercube multiprocessors. For an n x n matrix on p processors, the communication cost of this algorithm is O(n/[radical]p + log(p)), independent of the matrix sparsity pattern. The performance of the algorithm is demonstrated by employing it as the kernel in the well-known NAS conjugate gradient benchmark, where a run time of 6.09 seconds was observed. This is the best published performance on this benchmark achieved to date using a massively parallel supercomputer.

A secure file manager for UNIX

Energy Technology Data Exchange (ETDEWEB)

DeVries, R.G.

1990-12-31

The development of a secure file management system for a UNIX-based computer facility with supercomputers and workstations is described. Specifically, UNIX in its usual form does not address: (1) Operation which would satisfy rigorous security requirements. (2) Online space management in an environment where total data demands would be many times the actual online capacity. (3) Making the file management system part of a computer network in which users of any computer in the local network could retrieve data generated on any other computer in the network. The characteristics of UNIX can be exploited to develop a portable, secure file manager which would operate on computer systems ranging from workstations to supercomputers. Implementation considerations making unusual use of UNIX features, rather than requiring extensive internal system changes, are described, and implementation using the Cray Research Inc. UNICOS operating system is outlined.

The computational future for climate and Earth system models: on the path to petaflop and beyond.

Science.gov (United States)

Washington, Warren M; Buja, Lawrence; Craig, Anthony

2009-03-13

The development of the climate and Earth system models has had a long history, starting with the building of individual atmospheric, ocean, sea ice, land vegetation, biogeochemical, glacial and ecological model components. The early researchers were much aware of the long-term goal of building the Earth system models that would go beyond what is usually included in the climate models by adding interactive biogeochemical interactions. In the early days, the progress was limited by computer capability, as well as by our knowledge of the physical and chemical processes. Over the last few decades, there has been much improved knowledge, better observations for validation and more powerful supercomputer systems that are increasingly meeting the new challenges of comprehensive models. Some of the climate model history will be presented, along with some of the successes and difficulties encountered with present-day supercomputer systems.

Using a Cray Y-MP as an array processor for a RISC Workstation

Science.gov (United States)

Lamaster, Hugh; Rogallo, Sarah J.

1992-01-01

As microprocessors increase in power, the economics of centralized computing has changed dramatically. At the beginning of the 1980's, mainframes and super computers were often considered to be cost-effective machines for scalar computing. Today, microprocessor-based RISC (reduced-instruction-set computer) systems have displaced many uses of mainframes and supercomputers. Supercomputers are still cost competitive when processing jobs that require both large memory size and high memory bandwidth. One such application is array processing. Certain numerical operations are appropriate to use in a Remote Procedure Call (RPC)-based environment. Matrix multiplication is an example of an operation that can have a sufficient number of arithmetic operations to amortize the cost of an RPC call. An experiment which demonstrates that matrix multiplication can be executed remotely on a large system to speed the execution over that experienced on a workstation is described.

Scalable devices

KAUST Repository

Krü ger, Jens J.; Hadwiger, Markus

2014-01-01

In computer science in general and in particular the field of high performance computing and supercomputing the term scalable plays an important role. It indicates that a piece of hardware, a concept, an algorithm, or an entire system scales

76 FR 37100 - Intent To Prepare an Environmental Impact Statement; Continued Operation of the Department of...

Science.gov (United States)

2011-06-24

..., hazardous and radioactive material transportation, energy efficiency and renewable energy, nuclear energy, fossil energy, magnetic fusion, basic energy sciences, supercomputing, and biological and environmental.... Further, an updated evaluation of SNL/NM operational and transportation accident analyses and a new...

Parallel Computing and Model Evaluation for Environmental Systems: An Overview of the Supermuse and Frames Software Technologies

Science.gov (United States)

ERD’s Supercomputer for Model Uncertainty and Sensitivity Evaluation (SuperMUSE) is a key to enhancing quality assurance in environmental models and applications. Uncertainty analysis and sensitivity analysis remain critical, though often overlooked steps in the development and e...

A platform independent communication library for distributed computing

NARCIS (Netherlands)

Groen, D.; Rieder, S.; Grosso, P.; de Laat, C.; Portegies Zwart, S.

2010-01-01

We present MPWide, a platform independent communication library for performing message passing between supercomputers. Our library couples several local MPI applications through a long distance network using, for example, optical links. The implementation is deliberately kept light-weight, platform

Large-scale grid-enabled lattice-Boltzmann simulations of complex fluid flow in porous media and under shear

NARCIS (Netherlands)

Harting, J.D.R.; Venturoli, M.; Coveney, P.V.

2004-01-01

Well–designed lattice Boltzmann codes exploit the essentially embarrassingly parallel features of the algorithm and so can be run with considerable efficiency on modern supercomputers. Such scalable codes permit us to simulate the behaviour of increasingly large quantities of complex condensed

Scaling Techniques for Massive Scale-Free Graphs in Distributed (External) Memory

KAUST Repository

Pearce, Roger; Gokhale, Maya; Amato, Nancy M.

2013-01-01

We present techniques to process large scale-free graphs in distributed memory. Our aim is to scale to trillions of edges, and our research is targeted at leadership class supercomputers and clusters with local non-volatile memory, e.g., NAND Flash

Resonance – Journal of Science Education | News

Indian Academy of Sciences (India)

Programming Languages - A Brief Review. V Rajaraman ... V Rajaraman1 2. IBM Professor of Information Technology, Jawaharlal Nehru Centre for Advanced Scientific Research, Bangalore 560012, India; Hon.Professor, Supercomputer Education & Research Centre Indian Institute of Science, Bangalore 560012, India ...

The cat is out of the bag

KAUST Repository

Ananthanarayanan, Rajagopal; Esser, Steven K.; Simon, Horst D.; Modha, Dharmendra S.

2009-01-01

In the quest for cognitive computing, we have built a massively parallel cortical simulator, C2, that incorporates a number of innovations in computation, memory, and communication. Using C2 on LLNL's Dawn Blue Gene/P supercomputer with 147, 456

First fruits of LAA

International Nuclear Information System (INIS)

Anon.

1988-01-01

The LAA programme is divided into nine specialized areas - high precision tracking, calorimetry, large area devices, leading particle detection, special applications of integrated circuits, data acquisition and analysis, supercomputers and simulation studies, high magnetic fields, and high temperature superconductivity. (orig./HSI).

Multiscale aspects of the visual system and their use for scale invariant object recognition

NARCIS (Netherlands)

Petkov, N; vanDeemter, J; Karsch, F; Monien, B; Satz, H

1997-01-01

Psychophysical, neuroanatomical and neurophysiological evidence for multiscale aspects of the visual system is considered. The stack model and its relation to the image pyramid are discussed. The results of a straightforward implementation on a parallel supercomputer are presented. The high

Fellowship | Indian Academy of Sciences

Indian Academy of Sciences (India)

Rajaraman, Prof. Vaidyeswaran S.M. (MIT), Ph.D. (Wisconsin), FNA, FNAE, FNASc. Date of birth: 8 September 1933. Specialization: Computer Science and Information Systems Design Address: Honorary Professor, Supercomputer Education and Research Centre, Indian Institute of Science, Bengaluru 560 012, Karnataka

Update on the Culicoides sonorensis transcriptome project: a peek into the molecular biology of the midge

Science.gov (United States)

Next Generation Sequencing is transforming the way scientists collect and measure an organism’s genetic background and gene dynamics, while bioinformatics and super-computing are merging to facilitate parallel sample computation and interpretation at unprecedented speeds. Analyzing the complete gene...

NCEP Central Operations

Science.gov (United States)

) NCO Organizational Chart NOAA's Weather and Climate Operational Supercomputing System is known as Climate Climate Prediction Climate Archives Weather Safety Storm Ready NOAA Central Library Photo Library NCO's MISSION * Execute the NCEP operational model suite - Create climate, weather, ocean, space and

Climate@Home: Crowdsourcing Climate Change Research

Science.gov (United States)

Xu, C.; Yang, C.; Li, J.; Sun, M.; Bambacus, M.

2011-12-01

Climate change deeply impacts human wellbeing. Significant amounts of resources have been invested in building super-computers that are capable of running advanced climate models, which help scientists understand climate change mechanisms, and predict its trend. Although climate change influences all human beings, the general public is largely excluded from the research. On the other hand, scientists are eagerly seeking communication mediums for effectively enlightening the public on climate change and its consequences. The Climate@Home project is devoted to connect the two ends with an innovative solution: crowdsourcing climate computing to the general public by harvesting volunteered computing resources from the participants. A distributed web-based computing platform will be built to support climate computing, and the general public can 'plug-in' their personal computers to participate in the research. People contribute the spare computing power of their computers to run a computer model, which is used by scientists to predict climate change. Traditionally, only super-computers could handle such a large computing processing load. By orchestrating massive amounts of personal computers to perform atomized data processing tasks, investments on new super-computers, energy consumed by super-computers, and carbon release from super-computers are reduced. Meanwhile, the platform forms a social network of climate researchers and the general public, which may be leveraged to raise climate awareness among the participants. A portal is to be built as the gateway to the climate@home project. Three types of roles and the corresponding functionalities are designed and supported. The end users include the citizen participants, climate scientists, and project managers. Citizen participants connect their computing resources to the platform by downloading and installing a computing engine on their personal computers. Computer climate models are defined at the server side. Climate

Integration Of PanDA Workload Management System With Supercomputers

CERN Document Server

Klimentov, Alexei; The ATLAS collaboration; Maeno, Tadashi; Mashinistov, Ruslan; Nilsson, Paul; Oleynik, Danila; Panitkin, Sergey; Read, Kenneth; Ryabinkin, Evgeny; Wenaus, Torre

2015-01-01

The Large Hadron Collider (LHC), operating at the international CERN Laboratory in Geneva, Switzerland, is leading Big Data driven scientific explorations. Experiments at the LHC explore the fundamental nature of matter and the basic forces that shape our universe, and were recently credited for the discovery of a Higgs boson. ATLAS, one of the largest collaborations ever assembled in the sciences, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. The ATLAS experiment uses PanDA (Production and Data Analysis) Workload Management System for managing the workflow for all data processing on over 140 data centers. Through PanDA, ATLAS physicists see a single computing facility that enables rapid scientific breakthroughs for the experiment, even though the data centers are physically scattered all over the world. While PanDA currently uses more than 100,000 co...

Supercomputing for weather and climate modelling: convenience or necessity

CSIR Research Space (South Africa)

Landman, WA

2009-12-01

Full Text Available Weather and climate modelling require dedicated computer infrastructure in order to generate high-resolution, large ensemble, various models with different configurations, etc. in order to optimise operational forecasts and climate projections. High...

Study of ATLAS TRT performance with GRID and supercomputers.

CERN Document Server

Krasnopevtsev, Dimitriy; The ATLAS collaboration; Mashinistov, Ruslan; Belyaev, Nikita; Ryabinkin, Evgeny

2015-01-01

After the early success in discovering a new particle consistent with the long awaited Higgs boson, Large Hadron Collider experiments are ready for the precision measurements and further discoveries that will be made possible by much higher LHC collision rates from spring 2015. A proper understanding of the detectors performance at high occupancy conditions is important for many on-going physics analyses. The ATLAS Transition Radiation Tracker (TRT) is one of these detectors. TRT is a large straw tube tracking system that is the outermost of the three subsystems of the ATLAS Inner Detector (ID). TRT contributes significantly to the resolution for high-pT tracks in the ID providing excellent particle identification capabilities and electron-pion separation. ATLAS experiment is using Worldwide LHC Computing Grid. WLCG is a global collaboration of computer centers and provides seamless access to computing resources which include data storage capacity, processing power, sensors, visualisation tools and more. WLCG...

Study of ATLAS TRT performance with GRID and supercomputers.

CERN Document Server

Krasnopevtsev, Dimitriy; The ATLAS collaboration; Belyaev, Nikita; Mashinistov, Ruslan; Ryabinkin, Evgeny

2015-01-01

After the early success in discovering a new particle consistent with the long awaited Higgs boson, Large Hadron Collider experiments are ready for the precision measurements and further discoveries that will be made possible by much higher LHC collision rates from spring 2015. A proper understanding of the detectors performance at highoccupancy conditions is important for many on-going physics analyses. The ATLAS Transition Radiation Tracker (TRT) is one of these detectors. TRT is a large straw tube tracking system that is the outermost of the three subsystems of the ATLAS Inner Detector (ID). TRT contributes significantly to the resolution for high-pT tracks in the ID providing excellent particle identification capabilities and electron-pion separation. ATLAS experiment is using Worldwide LHC Computing Grid. WLCG is a global collaboration of computer centers and provides seamless access to computing resources which include data storage capacity, processing power, sensors, visualization tools and more. WLCG ...

Physics-Based Virtual Fly-Outs of Projectiles on Supercomputers

National Research Council Canada - National Science Library

Sahu, Jubaraj

2006-01-01

...) have been successfully fully coupled on high performance computing (HPC) platforms for Virtual Fly-Outs of guided munitions identical to actual free flight tests in the aerodynamic experimental facilities...

Efficient multitasking of Choleski matrix factorization on CRAY supercomputers

Science.gov (United States)

Overman, Andrea L.; Poole, Eugene L.

1991-01-01

A Choleski method is described and used to solve linear systems of equations that arise in large scale structural analysis. The method uses a novel variable-band storage scheme and is structured to exploit fast local memory caches while minimizing data access delays between main memory and vector registers. Several parallel implementations of this method are described for the CRAY-2 and CRAY Y-MP computers demonstrating the use of microtasking and autotasking directives. A portable parallel language, FORCE, is used for comparison with the microtasked and autotasked implementations. Results are presented comparing the matrix factorization times for three representative structural analysis problems from runs made in both dedicated and multi-user modes on both computers. CPU and wall clock timings are given for the parallel implementations and are compared to single processor timings of the same algorithm.

NREL Receives Editors' Choice Awards for Supercomputer Research | News |

Science.gov (United States)

performance data center, high-bay labs, and office space. NREL's Martha Symko-Davies honored by Women in successful women working in the energy field. As NREL's Director of Partnerships for Energy Systems awards for the Peregrine high-performance computer and the groundbreaking research it made possible. The

International Conference Nuclear Theory in the Supercomputing Era 2014

CERN Document Server

2014-01-01

The conference focuses on forefront challenges in physics, namely the fundamentals of nuclear structure and reactions, the origin of the strong inter-nucleon interactions from QCD, and computational nuclear physics with leadership class computer facilities to provide forefront simulations leading to new discoveries.This is the fourth in the series of NTSE-HITES conferences aimed to bring together nuclear theorists, computer scientists and applied mathematicians.

Interactive and Large Scale Supercomputing Simulations in Nonlinear Optics

National Research Council Canada - National Science Library

Moloney, J

2001-01-01

.... The upgrade consisted in purchasing 8 of the newest generation of 400 MHz CPUs, converting one of ONYX2 racks into a fully loaded 16-processor Origin 2000/2400 system and moving both high performance...

Multiscale Hy3S: Hybrid stochastic simulation for supercomputers

Directory of Open Access Journals (Sweden)

Kaznessis Yiannis N

2006-02-01

Full Text Available Abstract Background Stochastic simulation has become a useful tool to both study natural biological systems and design new synthetic ones. By capturing the intrinsic molecular fluctuations of "small" systems, these simulations produce a more accurate picture of single cell dynamics, including interesting phenomena missed by deterministic methods, such as noise-induced oscillations and transitions between stable states. However, the computational cost of the original stochastic simulation algorithm can be high, motivating the use of hybrid stochastic methods. Hybrid stochastic methods partition the system into multiple subsets and describe each subset as a different representation, such as a jump Markov, Poisson, continuous Markov, or deterministic process. By applying valid approximations and self-consistently merging disparate descriptions, a method can be considerably faster, while retaining accuracy. In this paper, we describe Hy3S, a collection of multiscale simulation programs. Results Building on our previous work on developing novel hybrid stochastic algorithms, we have created the Hy3S software package to enable scientists and engineers to both study and design extremely large well-mixed biological systems with many thousands of reactions and chemical species. We have added adaptive stochastic numerical integrators to permit the robust simulation of dynamically stiff biological systems. In addition, Hy3S has many useful features, including embarrassingly parallelized simulations with MPI; special discrete events, such as transcriptional and translation elongation and cell division; mid-simulation perturbations in both the number of molecules of species and reaction kinetic parameters; combinatorial variation of both initial conditions and kinetic parameters to enable sensitivity analysis; use of NetCDF optimized binary format to quickly read and write large datasets; and a simple graphical user interface, written in Matlab, to help users create biological systems and analyze data. We demonstrate the accuracy and efficiency of Hy3S with examples, including a large-scale system benchmark and a complex bistable biochemical network with positive feedback. The software itself is open-sourced under the GPL license and is modular, allowing users to modify it for their own purposes. Conclusion Hy3S is a powerful suite of simulation programs for simulating the stochastic dynamics of networks of biochemical reactions. Its first public version enables computational biologists to more efficiently investigate the dynamics of realistic biological systems.

AI, Native Supercomputing and The Revival of Moore's Law

OpenAIRE

Lu, Chien-Ping

2017-01-01

Based on Alan Turing's proposition on AI and computing machinery, which shaped Computing as we know it today, the new AI computing machinery should comprise a universal computer and a universal learning machine. The later should understand linear algebra natively to overcome the slowdown of Moore's law. In such a universal learnig machine, a computing unit does not need to keep the legacy of a universal computing core. The data can be distributed to the computing units, and the results can be...

High-performance modeling of CO2 sequestration by coupling reservoir simulation and molecular dynamics

KAUST Repository

Bao, Kai; Yan, Mi; Lu, Ligang; Allen, Rebecca; Salam, Amgad; Jordan, Kirk E.; Sun, Shuyu

2013-01-01

multicomponent compositional flow simulation to handle more complicated physical process in the future. Accuracy and scalability analysis are performed on an IBM BlueGene/P and on an IBM BlueGene/Q, the latest IBM supercomputer. Results show good accuracy of our

Know Your Personal Computer

Indian Academy of Sciences (India)

Home; Journals; Resonance – Journal of Science Education; Volume 1; Issue 11. Know Your Personal Computer 5. The CPU Base Instruction Set and Assembly Language Programming. Siddhartha Kumar Ghoshal ... Supercomputer Education and Research Centre, Indian Institute of Science, Bangalore 560 012, India ...

How AI turned the chess world upside down

KAUST Repository

Polgar, Susan

2018-01-01

Man vs. Machine! In 1977 an IBM supercomputer 'Deep Blue' beat the reigning World Chess Champion Gary Kasparov in a 6 game chess match played under tournament conditions. This was the first time a machine beat a human at chess. It symbolized

Discrete event simulation of NASA's Remote Exploration and Experimentation Project (REE)

Science.gov (United States)

Dunphy, J.; Rogstad, S.

2001-01-01

The Remote Exploration and Experimentation Project (REE) is a new initiative at JPL to be able to place a supercomputer on board a spacecraft and allow large amounts of data reduction and compression to be done before science results are returned to Earth.

All the world's a lattice

Energy Technology Data Exchange (ETDEWEB)

Pendlebooth, G. (Edinburgh Univ. (UK). Dept. of Physics)

1991-01-05

The use of supercomputers to generate lattice Monte Carlo simulations of the Standard Model in quantum mechanics is described. Researchers in the field of quantum chromodynamics need this method of testing theory against observation as paper calculation are too complex and difficult. (UK).

Power and thermal efficient numerical processing

DEFF Research Database (Denmark)

Liu, Wei; Nannarelli, Alberto

2015-01-01

Numerical processing is at the core of applications in many areas ranging from scientific and engineering calculations to financial computing. These applications are usually executed on large servers or supercomputers to exploit their high speed, high level of parallelism and high bandwidth...

Petascale Computational Systems

OpenAIRE

Bell, Gordon; Gray, Jim; Szalay, Alex

2007-01-01

Computational science is changing to be data intensive. Super-Computers must be balanced systems; not just CPU farms but also petascale IO and networking arrays. Anyone building CyberInfrastructure should allocate resources to support a balanced Tier-1 through Tier-3 design.

Object-Oriented Scientific Programming with Fortran 90

Science.gov (United States)

Norton, C.

1998-01-01

Fortran 90 is a modern language that introduces many important new features beneficial for scientific programming. We discuss our experiences in plasma particle simulation and unstructured adaptive mesh refinement on supercomputers, illustrating the features of Fortran 90 that support the object-oriented methodology.

Hierarchical approach to optimization of parallel matrix multiplication on large-scale platforms

KAUST Repository

Hasanov, Khalid

2014-03-04

© 2014, Springer Science+Business Media New York. Many state-of-the-art parallel algorithms, which are widely used in scientific applications executed on high-end computing systems, were designed in the twentieth century with relatively small-scale parallelism in mind. Indeed, while in 1990s a system with few hundred cores was considered a powerful supercomputer, modern top supercomputers have millions of cores. In this paper, we present a hierarchical approach to optimization of message-passing parallel algorithms for execution on large-scale distributed-memory systems. The idea is to reduce the communication cost by introducing hierarchy and hence more parallelism in the communication scheme. We apply this approach to SUMMA, the state-of-the-art parallel algorithm for matrix–matrix multiplication, and demonstrate both theoretically and experimentally that the modified Hierarchical SUMMA significantly improves the communication cost and the overall performance on large-scale platforms.

Overview of lattice gauge theory at the CSSM

International Nuclear Information System (INIS)

Williams, A.G.

2002-01-01

Full text: I present an overview of the lattice gauge theory effort at the Special Research Centre for the Subatomic Structure of Matter (CSSM). The CSSM specializes in research into the strong interactions and into quantum chromodynamics (QCD), which is the fundamental quantum gauge field theory of the strong interactions. The primary mission of the CSSM is to attempt to solve QCD and hence test the implications of the theory against experimental evidence. The difficulty lies in the fact that the QCD is a highly nonlinear, strongly coupled theory. The only known first-principles means to solve it is to approximate space-time by a four-dimensional 'grid' or 'lattice' and to simulate this 'lattice QCD' on massively parallel supercomputers. A discussion of the Orion supercomputer of the National Computing Facility for Lattice Gauge Theory (NFCLGT) and the latest QCD predictions obtained from Orion by CSSM researchers will be presented

Engineering-Based Thermal CFD Simulations on Massive Parallel Systems

KAUST Repository

Frisch, Jérôme

2015-05-22

The development of parallel Computational Fluid Dynamics (CFD) codes is a challenging task that entails efficient parallelization concepts and strategies in order to achieve good scalability values when running those codes on modern supercomputers with several thousands to millions of cores. In this paper, we present a hierarchical data structure for massive parallel computations that supports the coupling of a Navier–Stokes-based fluid flow code with the Boussinesq approximation in order to address complex thermal scenarios for energy-related assessments. The newly designed data structure is specifically designed with the idea of interactive data exploration and visualization during runtime of the simulation code; a major shortcoming of traditional high-performance computing (HPC) simulation codes. We further show and discuss speed-up values obtained on one of Germany’s top-ranked supercomputers with up to 140,000 processes and present simulation results for different engineering-based thermal problems.

Computational Approaches to Simulation and Optimization of Global Aircraft Trajectories

Science.gov (United States)

Ng, Hok Kwan; Sridhar, Banavar

2016-01-01

This study examines three possible approaches to improving the speed in generating wind-optimal routes for air traffic at the national or global level. They are: (a) using the resources of a supercomputer, (b) running the computations on multiple commercially available computers and (c) implementing those same algorithms into NASAs Future ATM Concepts Evaluation Tool (FACET) and compares those to a standard implementation run on a single CPU. Wind-optimal aircraft trajectories are computed using global air traffic schedules. The run time and wait time on the supercomputer for trajectory optimization using various numbers of CPUs ranging from 80 to 10,240 units are compared with the total computational time for running the same computation on a single desktop computer and on multiple commercially available computers for potential computational enhancement through parallel processing on the computer clusters. This study also re-implements the trajectory optimization algorithm for further reduction of computational time through algorithm modifications and integrates that with FACET to facilitate the use of the new features which calculate time-optimal routes between worldwide airport pairs in a wind field for use with existing FACET applications. The implementations of trajectory optimization algorithms use MATLAB, Python, and Java programming languages. The performance evaluations are done by comparing their computational efficiencies and based on the potential application of optimized trajectories. The paper shows that in the absence of special privileges on a supercomputer, a cluster of commercially available computers provides a feasible approach for national and global air traffic system studies.

Remote collaboration system based on large scale simulation

International Nuclear Information System (INIS)

Kishimoto, Yasuaki; Sugahara, Akihiro; Li, J.Q.

2008-01-01

Large scale simulation using super-computer, which generally requires long CPU time and produces large amount of data, has been extensively studied as a third pillar in various advanced science fields in parallel to theory and experiment. Such a simulation is expected to lead new scientific discoveries through elucidation of various complex phenomena, which are hardly identified only by conventional theoretical and experimental approaches. In order to assist such large simulation studies for which many collaborators working at geographically different places participate and contribute, we have developed a unique remote collaboration system, referred to as SIMON (simulation monitoring system), which is based on client-server system control introducing an idea of up-date processing, contrary to that of widely used post-processing. As a key ingredient, we have developed a trigger method, which transmits various requests for the up-date processing from the simulation (client) running on a super-computer to a workstation (server). Namely, the simulation running on a super-computer actively controls the timing of up-date processing. The server that has received the requests from the ongoing simulation such as data transfer, data analyses, and visualizations, etc. starts operations according to the requests during the simulation. The server makes the latest results available to web browsers, so that the collaborators can monitor the results at any place and time in the world. By applying the system to a specific simulation project of laser-matter interaction, we have confirmed that the system works well and plays an important role as a collaboration platform on which many collaborators work with one another

Application of Raptor-M3G to reactor dosimetry problems on massively parallel architectures - 026

International Nuclear Information System (INIS)

Longoni, G.

2010-01-01

The solution of complex 3-D radiation transport problems requires significant resources both in terms of computation time and memory availability. Therefore, parallel algorithms and multi-processor architectures are required to solve efficiently large 3-D radiation transport problems. This paper presents the application of RAPTOR-M3G (Rapid Parallel Transport Of Radiation - Multiple 3D Geometries) to reactor dosimetry problems. RAPTOR-M3G is a newly developed parallel computer code designed to solve the discrete ordinates (SN) equations on multi-processor computer architectures. This paper presents the results for a reactor dosimetry problem using a 3-D model of a commercial 2-loop pressurized water reactor (PWR). The accuracy and performance of RAPTOR-M3G will be analyzed and the numerical results obtained from the calculation will be compared directly to measurements of the neutron field in the reactor cavity air gap. The parallel performance of RAPTOR-M3G on massively parallel architectures, where the number of computing nodes is in the order of hundreds, will be analyzed up to four hundred processors. The performance results will be presented based on two supercomputing architectures: the POPLE supercomputer operated by the Pittsburgh Supercomputing Center and the Westinghouse computer cluster. The Westinghouse computer cluster is equipped with a standard Ethernet network connection and an InfiniBand R interconnects capable of a bandwidth in excess of 20 GBit/sec. Therefore, the impact of the network architecture on RAPTOR-M3G performance will be analyzed as well. (authors)

Migration of vectorized iterative solvers to distributed memory architectures

Energy Technology Data Exchange (ETDEWEB)

Pommerell, C. [AT& T Bell Labs., Murray Hill, NJ (United States); Ruehl, R. [CSCS-ETH, Manno (Switzerland)

1994-12-31

Both necessity and opportunity motivate the use of high-performance computers for iterative linear solvers. Necessity results from the size of the problems being solved-smaller problems are often better handled by direct methods. Opportunity arises from the formulation of the iterative methods in terms of simple linear algebra operations, even if this {open_quote}natural{close_quotes} parallelism is not easy to exploit in irregularly structured sparse matrices and with good preconditioners. As a result, high-performance implementations of iterative solvers have attracted a lot of interest in recent years. Most efforts are geared to vectorize or parallelize the dominating operation-structured or unstructured sparse matrix-vector multiplication, or to increase locality and parallelism by reformulating the algorithm-reducing global synchronization in inner products or local data exchange in preconditioners. Target architectures for iterative solvers currently include mostly vector supercomputers and architectures with one or few optimized (e.g., super-scalar and/or super-pipelined RISC) processors and hierarchical memory systems. More recently, parallel computers with physically distributed memory and a better price/performance ratio have been offered by vendors as a very interesting alternative to vector supercomputers. However, programming comfort on such distributed memory parallel processors (DMPPs) still lags behind. Here the authors are concerned with iterative solvers and their changing computing environment. In particular, they are considering migration from traditional vector supercomputers to DMPPs. Application requirements force one to use flexible and portable libraries. They want to extend the portability of iterative solvers rather than reimplementing everything for each new machine, or even for each new architecture.

Energy Secretary Dedicates ESIF at NREL | News | NREL

Science.gov (United States)

3 » Energy Secretary Dedicates ESIF at NREL Energy Secretary Dedicates ESIF at NREL September 18 prey. Enlarge image Energy Secretary Ernest Moniz (center) joins NREL Director Dan Arvizu (left) and newest Energy Department supercomputer. The high performance computer inside NREL's new Energy Systems

Comparison of some parallelization strategies of thermalhydraulic codes on GPUs

International Nuclear Information System (INIS)

Jendoubi, T.; Bergeaud, V.; Geay, A.

2013-01-01

Modern supercomputers architecture is now often based on hybrid concepts combining parallelism to distributed memory, parallelism to shared memory and also to GPUs (Graphic Process Units). In this work, we propose a new approach to take advantage of these graphic cards in thermohydraulics algorithms. (authors)

Journal of Earth System Science | Indian Academy of Sciences

Indian Academy of Sciences (India)

Operational mesoscale atmospheric dispersion prediction using a parallel computing cluster · C V Srinivas R Venkatesan N V Muralidharan Someshwar Das Hari Dass P Eswara Kumar · More Details Abstract Fulltext PDF. An operational atmospheric dispersion prediction system is implemented on a cluster supercomputer ...

Dynamic photonic lightpaths in the StarPlane network

NARCIS (Netherlands)

Grosso, P.; Marchal, D.; Maassen, J.; Bernier, E.; Xu, L.; de Laat, C.

2009-01-01

The StarPlane project enables users to dynamically control network photonic paths. Applications running on the Distributed ASCI Supercomputer (DAS-3) can manipulate wavelengths in the Dutch research and education network SURFnet6. The goal is to achieve fast switching times so that when the

Easy Access to HPC Resources through the Application GUI

KAUST Repository

van Waveren, Matthijs; Seif, Ahmed; Hassanein, Nasr; Moon, David; O'Byrnes, Niall; Clo, Alain M.; Murugan, Karthikeyan; Arena, Antonio

2016-01-01

and supercomputer that automates the transfer of input data and job submission and also the retrieval of results to the researcher’s local workstation. The innovation is that the user now submits his jobs from within the application GUI on his workstation, and does

Grids for Kids gives next-generation IT an early start

CERN Multimedia

2008-01-01

"Grids for Kids gives children a crash course in grid computing," explains co-organiser Anna Cook of the Enabling Grids for E-sciencE project. "We introduce them to concepts such as middleware, parallel processing and supercomputing, and give them opportunities for hands-on learning.

High performance computing on vector systems

CERN Document Server

Roller, Sabine

2008-01-01

Presents the developments in high-performance computing and simulation on modern supercomputer architectures. This book covers trends in hardware and software development in general and specifically the vector-based systems and heterogeneous architectures. It presents innovative fields like coupled multi-physics or multi-scale simulations.

Performance Evaluation of a SOA-based Rack-To-Rack Switch for Optical Interconnects Exploiting NRZ-DPSK

DEFF Research Database (Denmark)

Karinou, Fotini; Borkowski, Robert; Prince, Kamau

2012-01-01

We experimentally study the transmission performance of 10-Gb/s NRZ-DPSK through concatenated AWG MUX/DMUXs and SOAs employed in an optimized 64×64 optical supercomputer interconnect architecture. NRZ-DPSK offers 9-dB higher dynamic range compared to conventional IM/DD....

Streaming Parallel GPU Acceleration of Large-Scale filter-based Spiking Neural Networks

NARCIS (Netherlands)

L.P. Slazynski (Leszek); S.M. Bohte (Sander)

2012-01-01

htmlabstractThe arrival of graphics processing (GPU) cards suitable for massively parallel computing promises a↵ordable large-scale neural network simulation previously only available at supercomputing facil- ities. While the raw numbers suggest that GPUs may outperform CPUs by at least an order of

The Taverna workflow suite: designing and executing workflows of Web Services on the desktop, web or in the cloud

NARCIS (Netherlands)

Wolstencroft, K.; Haines, R.; Fellows, D.; Williams, A.; Withers, D.; Owen, S.; Soiland-Reyes, S.; Dunlop, I.; Nenadic, A.; Fisher, P.; Bhagat, J.; Belhajjame, K.; Bacall, F.; Hardisty, A.; Nieva de la Hidalga, A.; Balcazar Vargas, M.P.; Sufi, S.; Goble, C.

2013-01-01

The Taverna workflow tool suite (http://www.taverna.org.uk) is designed to combine distributed Web Services and/or local tools into complex analysis pipelines. These pipelines can be executed on local desktop machines or through larger infrastructure (such as supercomputers, Grids or cloud

Le CERN à la quête du GRID

CERN Multimedia

Perrin, Brigitte

2004-01-01

CERN is about to make endless calculations for its new particule accelerator; as there is no enough powerful super-computers, researchers decided to connect together thousands of PC all over the world: the GRID is born, soon it will be as popular as the WEB

A Comparison of Three Microkernels

NARCIS (Netherlands)

Tanenbaum, A.S.

The future of supercomputing lies in massively parallel computers. The nodes of these machines will need a different kind of operating system than current computers have. Many researchers in the field believe that microkernels provide the kind of functionality and performance required. In this paper

Advanced Modulation Techniques for High-Performance Computing Optical Interconnects

DEFF Research Database (Denmark)

Karinou, Fotini; Borkowski, Robert; Zibar, Darko

2013-01-01

We experimentally assess the performance of a 64 × 64 optical switch fabric used for ns-speed optical cell switching in supercomputer optical interconnects. More specifically, we study four alternative modulation formats and detection schemes, namely, 10-Gb/s nonreturn-to-zero differential phase-...

Energy and technology review

Energy Technology Data Exchange (ETDEWEB)

1984-03-01

The Lawrence Livermore National Laboratory publishes the Energy and Technology Review Monthly. This periodical reviews progress mode is selected programs at the laboratory. This issue includes articles on in-situ coal gasification, on chromosomal aberrations in human sperm, on high speed cell sorting and on supercomputers.

Michael F. Crowley | NREL

Science.gov (United States)

Pittsburgh Supercomputing Center, and at The Scripps Research Institute with David Case and Charles Brooks NIH-funded collaboration with Professor Charles L. Brooks III at the University of Michigan Areas of to determine source of twisting in hydrogen bonding patterns. This modeling will help to understand

Bulletin of Materials Science | Indian Academy of Sciences

Indian Academy of Sciences (India)

In this paper, several typical examples recently worked out by our research group are introduced to indicate that these methodologies are actually possible to be successfully used to predict materials properties before experiments based on the present day state-of-art supercomputing systems. It includes scientific ...

Computational Solutions for Today’s Navy: New Methods are Being Employed to Meet the Navy’s Changing Software-Development Environment

Science.gov (United States)

2008-03-01

software- development environment. ▶ Frank W. Bentrem, Ph.D., John T. Sample, Ph.D., and Michael M. Harris he Naval Research Labor - atory (NRL) is the...sonars (Through-the-Sensor technology), supercomputer generated numer- ical models, and historical/ clima - tological databases. It uses a vari- ety of

Slide 3

Indian Academy of Sciences (India)

All present day computers (classical computers) including Supercomputers, Parallel computers, use binary (0,1) logic, and all computations follow from this Yes/No answer. Many scientists have wondered if Quantum Mechanical systems can provide a new paradigm for computation. Feynman (1982) suggested that it might ...

Making Research Cyberinfrastructure a Strategic Choice

Science.gov (United States)

Hacker, Thomas J.; Wheeler, Bradley C.

2007-01-01

The commoditization of low-cost hardware has enabled even modest-sized laboratories and research projects to own their own "supercomputers." The authors argue that this local solution undermines rather than amplifies the research potential of scholars. CIOs, provosts, and research technologists should consider carefully an overall…

Energy and technology review

International Nuclear Information System (INIS)

1984-03-01

The Lawrence Livermore National Laboratory publishes the Energy and Technology Review Monthly. This periodical reviews progress mode is selected programs at the laboratory. This issue includes articles on in-situ coal gasification, on chromosomal aberrations in human sperm, on high speed cell sorting and on supercomputers

Space Transportation and the Computer Industry: Learning from the Past

Science.gov (United States)

Merriam, M. L.; Rasky, D.

2002-01-01

Since the space shuttle began flying in 1981, NASA has made a number of attempts to advance the state of the art in space transportation. In spite of billions of dollars invested, and several concerted attempts, no replacement for the shuttle is expected before 2010. Furthermore, the cost of access to space has dropped very slowly over the last two decades. On the other hand, the same two decades have seen dramatic progress in the computer industry. Computational speeds have increased by about a factor of 1000 and available memory, disk space, and network bandwidth has seen similar increases. At the same time, the cost of computing has dropped by about a factor of 10000. Is the space transportation problem simply harder? Or is there something to be learned from the computer industry? In looking for the answers, this paper reviews the early history of NASA's experience with supercomputers and NASA's visionary course change in supercomputer procurement strategy.

A parallel orbital-updating based plane-wave basis method for electronic structure calculations

International Nuclear Information System (INIS)

Pan, Yan; Dai, Xiaoying; Gironcoli, Stefano de; Gong, Xin-Gao; Rignanese, Gian-Marco; Zhou, Aihui

2017-01-01

Highlights: • Propose three parallel orbital-updating based plane-wave basis methods for electronic structure calculations. • These new methods can avoid the generating of large scale eigenvalue problems and then reduce the computational cost. • These new methods allow for two-level parallelization which is particularly interesting for large scale parallelization. • Numerical experiments show that these new methods are reliable and efficient for large scale calculations on modern supercomputers. - Abstract: Motivated by the recently proposed parallel orbital-updating approach in real space method , we propose a parallel orbital-updating based plane-wave basis method for electronic structure calculations, for solving the corresponding eigenvalue problems. In addition, we propose two new modified parallel orbital-updating methods. Compared to the traditional plane-wave methods, our methods allow for two-level parallelization, which is particularly interesting for large scale parallelization. Numerical experiments show that these new methods are more reliable and efficient for large scale calculations on modern supercomputers.

Structural Biology in the context of EGEE

CERN Document Server

García, D; Carazo, J M; Valverde, J R; Moscicki, J; Muraru, A

2007-01-01

Electron microscopy (EM) is a crucial technique, which allows Structural Biology researchers to characterize macromolecular assemblies in distinct functional states. Image processing in three dimensional EM (3D-EM) is used by a flourishing community (exemplarized by the EU funded 3D-EM NoE) and is characterized by voluminous data and large computing requirements, making this a problem well suited for Grid computing and the EGEE infrastructure. There are various steps in the 3D-EM refinement process that may benefit from Grid computing. To start with, large numbers of experimental images need to be averaged. Nowadays, typically tens of thousands of images are used, while future studies may routinely employ millions of images. Our group has been developing Xmipp, a package for single-particle 3D-EM image processing. Using Xmipp, the classification of 91,000 ribosome projections into 4 classes took more than 2500 CPU hours using the resources of the MareNostrum supercomputer at the Barcelona Supercomputing Centr...

Cyberinfrastructure for high energy physics in Korea

International Nuclear Information System (INIS)

Cho, Kihyeon; Kim, Hyunwoo; Jeung, Minho

2010-01-01

We introduce the hierarchy of cyberinfrastructure which consists of infrastructure (supercomputing and networks), Grid, e-Science, community and physics from bottom layer to top layer. KISTI is the national headquarter of supercomputer, network, Grid and e-Science in Korea. Therefore, KISTI is the best place to for high energy physicists to use cyberinfrastructure. We explain this concept on the CDF and the ALICE experiments. In the meantime, the goal of e-Science is to study high energy physics anytime and anywhere even if we are not on-site of accelerator laboratories. The components are data production, data processing and data analysis. The data production is to take both on-line and off-line shifts remotely. The data processing is to run jobs anytime, anywhere using Grid farms. The data analysis is to work together to publish papers using collaborative environment such as EVO (Enabling Virtual Organization) system. We also present the global community activities of FKPPL (France-Korea Particle Physics Laboratory) and physics as top layer.

Integration of the Chinese HPC Grid in ATLAS Distributed Computing

CERN Document Server

AUTHOR|(INSPIRE)INSPIRE-00081160; The ATLAS collaboration

2016-01-01

Fifteen Chinese High Performance Computing sites, many of them on the TOP500 list of most powerful supercomputers, are integrated into a common infrastructure providing coherent access to a user through an interface based on a RESTful interface called SCEAPI. These resources have been integrated into the ATLAS Grid production system using a bridge between ATLAS and SCEAPI which translates the authorization and job submission protocols between the two environments. The ARC Computing Element (ARC CE) forms the bridge using an extended batch system interface to allow job submission to SCEAPI. The ARC CE was setup at the Institute for High Energy Physics, Beijing, in order to be as close as possible to the SCEAPI front-end interface at the Computing Network Information Center, also in Beijing. This paper describes the technical details of the integration between ARC CE and SCEAPI and presents results so far with two supercomputer centers, Tianhe-IA and ERA. These two centers have been the pilots for ATLAS Monte C...

Integration of the Chinese HPC Grid in ATLAS Distributed Computing

CERN Document Server

AUTHOR|(INSPIRE)INSPIRE-00081160

2017-01-01

Fifteen Chinese High-Performance Computing sites, many of them on the TOP500 list of most powerful supercomputers, are integrated into a common infrastructure providing coherent access to a user through an interface based on a RESTful interface called SCEAPI. These resources have been integrated into the ATLAS Grid production system using a bridge between ATLAS and SCEAPI which translates the authorization and job submission protocols between the two environments. The ARC Computing Element (ARC-CE) forms the bridge using an extended batch system interface to allow job submission to SCEAPI. The ARC-CE was setup at the Institute for High Energy Physics, Beijing, in order to be as close as possible to the SCEAPI front-end interface at the Computing Network Information Center, also in Beijing. This paper describes the technical details of the integration between ARC-CE and SCEAPI and presents results so far with two supercomputer centers, Tianhe-IA and ERA. These two centers have been the pilots for ATLAS Monte C...

Fluctuations and transport in fusion plasmas. Final report

International Nuclear Information System (INIS)

Gould, R.W.; Liewer, P.C.

1995-01-01

The energy confinement in tokamaks in thought to be limited by transport caused by plasma turbulence. Three dimensional plasma particle-in-cell (PIC) codes are used to model the turbulent transport in tokamaks to attempt to understand this phenomena so that tokamaks can be made more efficient. Presently, hundreds of hours of Cray time are used to model these experiments and much bigger and longer runs are desired, to model a large tokamak with realistic parameters is beyond the capability of existing sequential supercomputers. Parallel supercomputers might be a cost effect tool for performing such large scale 3D tokamak simulations. The goal of the work was to develop algorithms for performing PIC codes on coarse-grained message passing parallel computers and to evaluate the performance of such parallel computers on PIC codes. This algorithm would be used in a large scale PIC production code such as the UCLA 3D gyrokinetic code

Science & Technology Review June 2012

Energy Technology Data Exchange (ETDEWEB)

Poyneer, L A

2012-04-20

This month's issue has the following articles: (1) A New Era in Climate System Analysis - Commentary by William H. Goldstein; (2) Seeking Clues to Climate Change - By comparing past climate records with results from computer simulations, Livermore scientists can better understand why Earth's climate has changed and how it might change in the future; (3) Finding and Fixing a Supercomputer's Faults - Livermore experts have developed innovative methods to detect hardware faults in supercomputers and help applications recover from errors that do occur; (4) Targeting Ignition - Enhancements to the cryogenic targets for National Ignition Facility experiments are furthering work to achieve fusion ignition with energy gain; (5) Neural Implants Come of Age - A new generation of fully implantable, biocompatible neural prosthetics offers hope to patients with neurological impairment; and (6) Incubator Busy Growing Energy Technologies - Six collaborations with industrial partners are using the Laboratory's high-performance computing resources to find solutions to urgent energy-related problems.

High Performance Computing in Science and Engineering '08 : Transactions of the High Performance Computing Center

CERN Document Server

Kröner, Dietmar; Resch, Michael

2009-01-01

The discussions and plans on all scienti?c, advisory, and political levels to realize an even larger “European Supercomputer” in Germany, where the hardware costs alone will be hundreds of millions Euro – much more than in the past – are getting closer to realization. As part of the strategy, the three national supercomputing centres HLRS (Stuttgart), NIC/JSC (Julic ¨ h) and LRZ (Munich) have formed the Gauss Centre for Supercomputing (GCS) as a new virtual organization enabled by an agreement between the Federal Ministry of Education and Research (BMBF) and the state ministries for research of Baden-Wurttem ¨ berg, Bayern, and Nordrhein-Westfalen. Already today, the GCS provides the most powerful high-performance computing - frastructure in Europe. Through GCS, HLRS participates in the European project PRACE (Partnership for Advances Computing in Europe) and - tends its reach to all European member countries. These activities aligns well with the activities of HLRS in the European HPC infrastructur...

DEISA2: supporting and developing a European high-performance computing ecosystem

International Nuclear Information System (INIS)

Lederer, H

2008-01-01

The DEISA Consortium has deployed and operated the Distributed European Infrastructure for Supercomputing Applications. Through the EU FP7 DEISA2 project (funded for three years as of May 2008), the consortium is continuing to support and enhance the distributed high-performance computing infrastructure and its activities and services relevant for applications enabling, operation, and technologies, as these are indispensable for the effective support of computational sciences for high-performance computing (HPC). The service-provisioning model will be extended from one that supports single projects to one supporting virtual European communities. Collaborative activities will also be carried out with new European and other international initiatives. Of strategic importance is cooperation with the PRACE project, which is preparing for the installation of a limited number of leadership-class Tier-0 supercomputers in Europe. The key role and aim of DEISA will be to deliver a turnkey operational solution for a persistent European HPC ecosystem that will integrate national Tier-1 centers and the new Tier-0 centers

Hybrid petacomputing meets cosmology: The Roadrunner Universe project

International Nuclear Information System (INIS)

Habib, Salman; Pope, Adrian; Lukic, Zarija; Daniel, David; Fasel, Patricia; Desai, Nehal; Heitmann, Katrin; Hsu, Chung-Hsing; Ankeny, Lee; Mark, Graham; Bhattacharya, Suman; Ahrens, James

2009-01-01

The target of the Roadrunner Universe project at Los Alamos National Laboratory is a set of very large cosmological N-body simulation runs on the hybrid supercomputer Roadrunner, the world's first petaflop platform. Roadrunner's architecture presents opportunities and difficulties characteristic of next-generation supercomputing. We describe a new code designed to optimize performance and scalability by explicitly matching the underlying algorithms to the machine architecture, and by using the physics of the problem as an essential aid in this process. While applications will differ in specific exploits, we believe that such a design process will become increasingly important in the future. The Roadrunner Universe project code, MC 3 (Mesh-based Cosmology Code on the Cell), uses grid and direct particle methods to balance the capabilities of Roadrunner's conventional (Opteron) and accelerator (Cell BE) layers. Mirrored particle caches and spectral techniques are used to overcome communication bandwidth limitations and possible difficulties with complicated particle-grid interaction templates.

Upside to downsizing : Acceleware's graphic processor technology propels seismic data processing revolution

Energy Technology Data Exchange (ETDEWEB)

Smith, M.

2009-11-15

Accelware has developed a graphic processor technology (GPU) that is transforming the petroleum industry. The benefits of the technology are its small-footprint, low-wattage, and high speed. The software brings supercomputing speed to the desktop by leveraging the massive parallel processing capacity to the very latest in GPU technology. This article discussed the GPU technology and its emergence as a powerful supercomputing tool. Accelware's partnering with California-based NVIDIA was also outlined. The advantages of the technology were also discussed including its smaller footprint. Accelware's hardware takes up a fraction of the space and uses up to 70 per cent less power than a traditional central processing unit. By combining Accelware's core knowledge in making complex algorithms run in parallel with an in-house team of seismic industry experts, the company provides software solutions for seismic data processors that access the massively parallel processing capabilities of GPUs. 1 fig.

Optimizing the Performance of Reactive Molecular Dynamics Simulations for Multi-core Architectures

Energy Technology Data Exchange (ETDEWEB)

Aktulga, Hasan Metin [Michigan State Univ., East Lansing, MI (United States); Coffman, Paul [Argonne National Lab. (ANL), Argonne, IL (United States); Shan, Tzu-Ray [Sandia National Lab. (SNL-NM), Albuquerque, NM (United States); Knight, Chris [Argonne National Lab. (ANL), Argonne, IL (United States); Jiang, Wei [Argonne National Lab. (ANL), Argonne, IL (United States)

2015-12-01

Hybrid parallelism allows high performance computing applications to better leverage the increasing on-node parallelism of modern supercomputers. In this paper, we present a hybrid parallel implementation of the widely used LAMMPS/ReaxC package, where the construction of bonded and nonbonded lists and evaluation of complex ReaxFF interactions are implemented efficiently using OpenMP parallelism. Additionally, the performance of the QEq charge equilibration scheme is examined and a dual-solver is implemented. We present the performance of the resulting ReaxC-OMP package on a state-of-the-art multi-core architecture Mira, an IBM BlueGene/Q supercomputer. For system sizes ranging from 32 thousand to 16.6 million particles, speedups in the range of 1.5-4.5x are observed using the new ReaxC-OMP software. Sustained performance improvements have been observed for up to 262,144 cores (1,048,576 processes) of Mira with a weak scaling efficiency of 91.5% in larger simulations containing 16.6 million particles.

Development of the real time monitor system

Energy Technology Data Exchange (ETDEWEB)

Kato, Katsumi [Research Organization for Information Science and Technology, Tokai, Ibaraki (Japan); Watanabe, Tadashi; Kaburaki, Hideo

1996-10-01

Large-scale simulation technique is studied at the Center for Promotion of Computational Science and Engineering (CCSE) for the computational science research in nuclear fields. Visualization and animation processing technique are studied and developed for efficient understanding of simulation results. The real time monitor system, in which on-going simulation results are transferred from a supercomputer or workstation to a graphic workstation and are visualized and recorded, is described in this report. This system is composed of the graphic workstation and the video equipment connected to the network. The control shell programs are the job-execution shell for simulations on supercomputers, the file-transfer shell for output files for visualization, and the shell for starting visualization tools. Special image processing technique and hardware are not necessary in this system and the standard visualization tool AVS and the UNIX commands are used, so that this system can be implemented and applied in various computer environments. (author)

Data decomposition of Monte Carlo particle transport simulations via tally servers

International Nuclear Information System (INIS)

Romano, Paul K.; Siegel, Andrew R.; Forget, Benoit; Smith, Kord

2013-01-01

An algorithm for decomposing large tally data in Monte Carlo particle transport simulations is developed, analyzed, and implemented in a continuous-energy Monte Carlo code, OpenMC. The algorithm is based on a non-overlapping decomposition of compute nodes into tracking processors and tally servers. The former are used to simulate the movement of particles through the domain while the latter continuously receive and update tally data. A performance model for this approach is developed, suggesting that, for a range of parameters relevant to LWR analysis, the tally server algorithm should perform with minimal overhead on contemporary supercomputers. An implementation of the algorithm in OpenMC is then tested on the Intrepid and Titan supercomputers, supporting the key predictions of the model over a wide range of parameters. We thus conclude that the tally server algorithm is a successful approach to circumventing classical on-node memory constraints en route to unprecedentedly detailed Monte Carlo reactor simulations

High performance computing system in the framework of the Higgs boson studies

CERN Document Server

Belyaev, Nikita; The ATLAS collaboration; Velikhov, Vasily; Konoplich, Rostislav

2017-01-01

The Higgs boson physics is one of the most important and promising fields of study in the modern high energy physics. It is important to notice, that GRID computing resources become strictly limited due to increasing amount of statistics, required for physics analyses and unprecedented LHC performance. One of the possibilities to address the shortfall of computing resources is the usage of computer institutes' clusters, commercial computing resources and supercomputers. To perform precision measurements of the Higgs boson properties in these realities, it is also highly required to have effective instruments to simulate kinematic distributions of signal events. In this talk we give a brief description of the modern distribution reconstruction method called Morphing and perform few efficiency tests to demonstrate its potential. These studies have been performed on the WLCG and Kurchatov Institute’s Data Processing Center, including Tier-1 GRID site and supercomputer as well. We also analyze the CPU efficienc...

Microcontroller-based network for meteorological sensing and weather forecast calculations

Directory of Open Access Journals (Sweden)

A. Vas

2012-06-01

Full Text Available Weather forecasting needs a lot of computing power. It is generally accomplished by using supercomputers which are expensive to rent and to maintain. In addition, weather services also have to maintain radars, balloons and pay for worldwide weather data measured by stations and satellites. Weather forecasting computations usually consist of solving differential equations based on the measured parameters. To do that, the computer uses the data of close and distant neighbor points. Accordingly, if small-sized weather stations, which are capable of making measurements, calculations and communication, are connected through the Internet, then they can be used to run weather forecasting calculations like a supercomputer does. It doesn’t need any central server to achieve this, because this network operates as a distributed system. We chose Microchip’s PIC18 microcontroller (μC platform in the implementation of the hardware, and the embedded software uses the TCP/IP Stack v5.41 provided by Microchip.

NREL Research Earns Two Prestigious R&D 100 Awards | News | NREL

Science.gov (United States)

and development not only create jobs in America but help advance the goal of a clean energy future and Steve Johnston. High Performance Supercomputing Platform Uses Warm Water to Prevent Heat Build-up initiative were NREL's Steve Hammond and Nicolas Dube of HP. "Oscars" of Innovation Winners of the

Resonance – Journal of Science Education | Indian Academy of ...

Indian Academy of Sciences (India)

Jatan K Modi1 Sachin P Nanavati2 Amit S Phadke1 Prasanta K Panigrahi3. Dharmsinh Desai Institute of Technology, Nadiad 387001, India. National PARAM Supercomputing FaCility, Centre for Development of Advanced Computing (C-DACI. Pune University Campus, Ganesh Khind, Pune 411 007, India. Physical ...

LUCKY-TD code for solving the time-dependent transport equation with the use of parallel computations

Energy Technology Data Exchange (ETDEWEB)

Moryakov, A. V., E-mail: sailor@orc.ru [National Research Centre Kurchatov Institute (Russian Federation)

2016-12-15

An algorithm for solving the time-dependent transport equation in the P{sub m}S{sub n} group approximation with the use of parallel computations is presented. The algorithm is implemented in the LUCKY-TD code for supercomputers employing the MPI standard for the data exchange between parallel processes.

Simulating the universe on an intercontinental grid

NARCIS (Netherlands)

Portegies Zwart, S.; Ishiyama, T.; Groen, D.; Nitadori, K.; Makino, J.; de Laat, C.; McMillan, S.; Hiraki, K.; Harfst, S.; Grosso, P.

2010-01-01

The computational requirements of simulating a sector of the universe led an international team of researchers to try concurrent processing on two supercomputers half a world apart. Data traveled nearly 27,000 km in 0.277 second, crisscrossing two oceans to go from Amsterdam to Tokyo and back.

Alleviating Search Uncertainty through Concept Associations: Automatic Indexing, Co-Occurrence Analysis, and Parallel Computing.

Science.gov (United States)

Chen, Hsinchun; Martinez, Joanne; Kirchhoff, Amy; Ng, Tobun D.; Schatz, Bruce R.

1998-01-01

Grounded on object filtering, automatic indexing, and co-occurrence analysis, an experiment was performed using a parallel supercomputer to analyze over 400,000 abstracts in an INSPEC computer engineering collection. A user evaluation revealed that system-generated thesauri were better than the human-generated INSPEC subject thesaurus in concept…

High performance adaptive image processing on multi-scale hybrid architectures

NARCIS (Netherlands)

Liu, Fangbin

2015-01-01

In such an exciting age of information explosion, huge amount of visual data are produced continuously 24 hours, 7 days in both daily life and scientific research. Processing and storage of such a huge amount of data forms big challenges. Use of supercomputers tackles the need-for-speed challenge

Internet, maratona di supercalcolo con Grid

CERN Multimedia

2005-01-01

At CERN, stream of data medium 600 megabytes per second during 10 consecutive days; the supercomputer Grid made a success its first challenge: eight supercalculation centers maintained a continuous flow of data coming from Cern and going to seven differents centers in Europe and the USA (1½ page)

Qualities of Grid Computing that can last for Ages | Asagba | Journal ...

African Journals Online (AJOL)

Grid computing has emerged as an important new field, distinguished from conventional distributed computing based on its abilities on large-scale resource sharing and services. And it will even become more popular because of the benefits it can offer over the traditional supercomputers, and other forms of distributed ...

Land degradation and property regimes

Science.gov (United States)

Paul M. Beaumont; Robert T. Walker

1996-01-01

This paper addresses the relationship between property regimes and land degradation outcomes, in the context of peasant agriculture. We consider explicitly whether private property provides for superior soil resource conservation, as compared to common property and open access. To assess this we implement optimization algorithms on a supercomputer to address resource...

Lattice overview

International Nuclear Information System (INIS)

Creutz, M.

1984-01-01

After reviewing some recent developments in supercomputer access, the author discusses a few areas where perturbation theory and lattice gauge simulations make contact. The author concludes with a brief discussion of a deterministic dynamics for the Ising model. This may be useful for numerical studies of nonequilibrium phenomena. 13 references

DICE/ColDICE: 6D collisionless phase space hydrodynamics using a lagrangian tesselation

Science.gov (United States)

Sousbie, Thierry

2018-01-01

DICE is a C++ template library designed to solve collisionless fluid dynamics in 6D phase space using massively parallel supercomputers via an hybrid OpenMP/MPI parallelization. ColDICE, based on DICE, implements a cosmological and physical VLASOV-POISSON solver for cold systems such as dark matter (CDM) dynamics.

Extended Subject Access to Hypertext Online Documentation. Parts I and II: The Search-Support and Maintenance Problems.

Science.gov (United States)

Girill, T. R.; And Others

1991-01-01

Describes enhancements made to a hypertext information retrieval system at the National Energy Research Supercomputer Center (NERSC) called DFT (Document, Find, and Theseus). The enrichment of DFT's entry vocabulary is described, DFT and other hypertext systems are compared, and problems that occur due to the need for frequent updates are…

Evaluating Satellite and Supercomputing Technologies for Improved Coastal Ecosystem Assessments

Science.gov (United States)

McCarthy, Matthew James

Water quality and wetlands represent two vital elements of a healthy coastal ecosystem. Both experienced substantial declines in the U.S. during the 20th century. Overall coastal wetland cover decreased over 50% in the 20th century due to coastal development and water pollution. Management and legislative efforts have successfully addressed some of the problems and threats, but recent research indicates that the diffuse impacts of climate change and non-point source pollution may be the primary drivers of current and future water-quality and wetland stress. In order to respond to these pervasive threats, traditional management approaches need to adopt modern technological tools for more synoptic, frequent and fine-scale monitoring and assessment. In this dissertation, I explored some of the applications possible with new, commercial satellite imagery to better assess the status of coastal ecosystems. Large-scale land-cover change influences the quality of adjacent coastal water. Satellite imagery has been used to derive land-cover maps since the 1960's. It provides multiple data points with which to evaluate the effects of land-cover change on water quality. The objective of the first chapter of this research was to determine how 40 years of land-cover change in the Tampa Bay watershed (6,500 km2) may have affected turbidity and chlorophyll concentration - two proxies for coastal water quality. Land cover classes were evaluated along with precipitation and wind stress as explanatory variables. Results varied between analyses for the entire estuary and those of segments within the bay. Changes in developed land percent cover best explained the turbidity and chlorophyll-concentration time series for the entire bay (R2 > 0.75, p Ocean-color satellite imagery was used to derive proxies for coastal water with near-daily satellite observations since 2000. The goal of chapter two was to identify drivers of turbidity variability for 11 National Estuary Program water bodies along the Gulf of Mexico. Land cover assessments could not be used as an explanatory variable because of the low temporal resolution (i.e. approximately one map per five-year period). Ocean color metrics were evaluated against atmospheric, meteorological, and oceanographic variables including precipitation, wind speed, U and V wind vectors, river discharge, and water level over weekly, monthly, seasonal and annual time steps. Climate indices like the North Atlantic Oscillation and El Nino Southern Oscillation index were also examined as possible drivers of long-term changes. Extreme turbidity events were defined by the 90th and 95th percentile observations over each time step. Wind speed, river discharge and El Nino best explained variability in turbidity time-series and extreme events (R2 > 0.2, p Oceanic and Atmospheric Administration, Southwest Florida Water Management District, and National Wetland Inventory, we found that these historical land cover products overestimated by 2-10 times the actual extent of wetlands as identified in the WorldView-2 maps. We could find no study that had utilized more than six of these commercial images for a given project. Part of the problem is cost of the images, but there is also the cost of processing the images, which is typically done one at a time and with substantial human interaction. Chapter four explains an approach to automate the preprocessing and classification of imagery to detect wetlands within the Tampa Bay watershed (6,500 km2). Software scripts in Python, Matlab and Linux were used to ingest 130 WorldView-2 images and to generate maps that included wetlands, uplands, water, and bare and developed land. These maps proved to be more accurate at identifying forested wetland (78%) than those by NOAA, SWFWMD, and NWI (45-65%) based on ground validation data. Typical processing methods would have required 4-5 months to complete this work, but this protocol completed the 130 images in under 24 hours. Chapter five of the dissertation reviews coastal management case studies that have used satellite technologies. The objective was to illustrate the utility of this technology. The management sectors reviewed included coral reefs, wetlands, water quality, public health, and fisheries and aquaculture.

UbiWorld: An environment integrating virtual reality, supercomputing, and design

Energy Technology Data Exchange (ETDEWEB)

Disz, T.; Papka, M.E.; Stevens, R. [Argonne National Lab., IL (United States). Mathematics and Computer Science Div.

1997-07-01

UbiWorld is a concept being developed by the Futures Laboratory group at Argonne National Laboratory that ties together the notion of ubiquitous computing (Ubicomp) with that of using virtual reality for rapid prototyping. The goal is to develop an environment where one can explore Ubicomp-type concepts without having to build real Ubicomp hardware. The basic notion is to extend object models in a virtual world by using distributed wide area heterogeneous computing technology to provide complex networking and processing capabilities to virtual reality objects.

Benchmarking and tuning the MILC code on clusters and supercomputers

International Nuclear Information System (INIS)

Gottlieb, Steven

2002-01-01

Recently, we have benchmarked and tuned the MILC code on a number of architectures including Intel Itanium and Pentium IV (PIV), dual-CPU Athlon, and the latest Compaq Alpha nodes. Results will be presented for many of these, and we shall discuss some simple code changes that can result in a very dramatic speedup of the KS conjugate gradient on processors with more advanced memory systems such as PIV, IBM SP and Alpha

Benchmarking and tuning the MILC code on clusters and supercomputers

International Nuclear Information System (INIS)

Steven A. Gottlieb

2001-01-01

Recently, we have benchmarked and tuned the MILC code on a number of architectures including Intel Itanium and Pentium IV (PIV), dual-CPU Athlon, and the latest Compaq Alpha nodes. Results will be presented for many of these, and we shall discuss some simple code changes that can result in a very dramatic speedup of the KS conjugate gradient on processors with more advanced memory systems such as PIV, IBM SP and Alpha

Benchmarking and tuning the MILC code on clusters and supercomputers

Science.gov (United States)

Gottlieb, Steven

2002-03-01

Recently, we have benchmarked and tuned the MILC code on a number of architectures including Intel Itanium and Pentium IV (PIV), dual-CPU Athlon, and the latest Compaq Alpha nodes. Results will be presented for many of these, and we shall discuss some simple code changes that can result in a very dramatic speedup of the KS conjugate gradient on processors with more advanced memory systems such as PIV, IBM SP and Alpha.

A visualization environment for supercomputing-based applications in computational mechanics

Energy Technology Data Exchange (ETDEWEB)

Pavlakos, C.J.; Schoof, L.A.; Mareda, J.F.

1993-06-01

In this paper, we characterize a visualization environment that has been designed and prototyped for a large community of scientists and engineers, with an emphasis in superconducting-based computational mechanics. The proposed environment makes use of a visualization server concept to provide effective, interactive visualization to the user`s desktop. Benefits of using the visualization server approach are discussed. Some thoughts regarding desirable features for visualization server hardware architectures are also addressed. A brief discussion of the software environment is included. The paper concludes by summarizing certain observations which we have made regarding the implementation of such visualization environments.

Petascale supercomputing to accelerate the design of high-temperature alloys

Science.gov (United States)

Shin, Dongwon; Lee, Sangkeun; Shyam, Amit; Haynes, J. Allen

2017-12-01

Recent progress in high-performance computing and data informatics has opened up numerous opportunities to aid the design of advanced materials. Herein, we demonstrate a computational workflow that includes rapid population of high-fidelity materials datasets via petascale computing and subsequent analyses with modern data science techniques. We use a first-principles approach based on density functional theory to derive the segregation energies of 34 microalloying elements at the coherent and semi-coherent interfaces between the aluminium matrix and the θ‧-Al2Cu precipitate, which requires several hundred supercell calculations. We also perform extensive correlation analyses to identify materials descriptors that affect the segregation behaviour of solutes at the interfaces. Finally, we show an example of leveraging machine learning techniques to predict segregation energies without performing computationally expensive physics-based simulations. The approach demonstrated in the present work can be applied to any high-temperature alloy system for which key materials data can be obtained using high-performance computing.

The Erasmus Computing Grid - Building a Super-Computer for FREE

NARCIS (Netherlands)

T.A. Knoch (Tobias); L.V. de Zeeuw (Luc)

2007-01-01

textabstractToday advances in scientific research as well as clinical diagnostics and treatment are inevitably connected with information solutions concerning computation power and information storage. The needs for information technology are enormous and are in many cases the limiting

Monte Carlo simulations of quantum systems on massively parallel supercomputers

International Nuclear Information System (INIS)

Ding, H.Q.

1993-01-01

A large class of quantum physics applications uses operator representations that are discrete integers by nature. This class includes magnetic properties of solids, interacting bosons modeling superfluids and Cooper pairs in superconductors, and Hubbard models for strongly correlated electrons systems. This kind of application typically uses integer data representations and the resulting algorithms are dominated entirely by integer operations. The authors implemented an efficient algorithm for one such application on the Intel Touchstone Delta and iPSC/860. The algorithm uses a multispin coding technique which allows significant data compactification and efficient vectorization of Monte Carlo updates. The algorithm regularly switches between two data decompositions, corresponding naturally to different Monte Carlo updating processes and observable measurements such that only nearest-neighbor communications are needed within a given decomposition. On 128 nodes of Intel Delta, this algorithm updates 183 million spins per second (compared to 21 million on CM-2 and 6.2 million on a Cray Y-MP). A systematic performance analysis shows a better than 90% efficiency in the parallel implementation

Performance analysis of job scheduling policies in parallel supercomputing environments

Energy Technology Data Exchange (ETDEWEB)

Naik, V.K.; Squillante, M.S. [IBM T.J. Watson Research Center, Yorktown Heights, NY (United States); Setia, S.K. [George Mason Univ., Fairfax, VA (United States). Dept. of Computer Science

1993-12-31

In this paper the authors analyze three general classes of scheduling policies under a workload typical of largescale scientific computing. These policies differ in the manner in which processors are partitioned among the jobs as well as the way in which jobs are prioritized for execution on the partitions. Their results indicate that existing static schemes do not perform well under varying workloads. Adaptive policies tend to make better scheduling decisions, but their ability to adjust to workload changes is limited. Dynamic partitioning policies, on the other hand, yield the best performance and can be tuned to provide desired performance differences among jobs with varying resource demands.

[Supercomputer investigation of the protein-ligand system low-energy minima].

Science.gov (United States)

Oferkin, I V; Sulimov, A V; Katkova, E V; Kutov, D K; Grigoriev, F V; Kondakova, O A; Sulimov, V B

2015-01-01

The accuracy of the protein-ligand binding energy calculations and ligand positioning is strongly influenced by the choice of the docking target function. This work demonstrates the evaluation of the five different target functions used in docking: functions based on MMFF94 force field and functions based on PM7 quantum-chemical method accounting or without accounting the implicit solvent model (PCM, COSMO or SGB). For these purposes the ligand positions corresponding to the minima of the target function and the experimentally known ligand positions in the protein active site (crystal ligand positions) were compared. Each function was examined on the same test-set of 16 protein-ligand complexes. The new parallelized docking program FLM based on Monte Carlo search algorithm was developed to perform the comprehensive low-energy minima search and to calculate the protein-ligand binding energy. This study demonstrates that the docking target function based on the MMFF94 force field can be used to detect the crystal or near crystal positions of the ligand by the finding the low-energy local minima spectrum of the target function. The importance of solvent accounting in the docking process for the accurate ligand positioning is also shown. The accuracy of the ligand positioning as well as the correlation between the calculated and experimentally determined protein-ligand binding energies are improved when the MMFF94 force field is substituted by the new PM7 method with implicit solvent accounting.

Lisbon: Supercomputer for Portugal financed from 'CERN Fund'

International Nuclear Information System (INIS)

Anon.

1990-01-01

A powerful new computer is now in use at the Portuguese National Foundation for Scientific Computation (FCCN Lisbon), set up in 1987 to help fund university computing, to anticipate future requirements and to provide a fast computer at the National Civil Engineering Laboratory (LNEC) as a central node for remote access by major research institutes

When Rural Reality Goes Virtual.

Science.gov (United States)

Husain, Dilshad D.

1998-01-01

In rural towns where sparse population and few business are barriers, virtual reality may be the only way to bring work-based learning to students. A partnership between a small-town high school, the Ohio Supercomputer Center, and a high-tech business will enable students to explore the workplace using virtual reality. (JOW)

VR system CompleXcope programming guide

International Nuclear Information System (INIS)

Kageyama, Akira; Sato, Tetsuya

1998-09-01

A CAVE virtual reality system CompleXcope is installed in Theory and Computer Center, National Institute for Fusion Science, for the purpose of the interactive analysis/visualization of 3-dimensional complex data of supercomputer simulations. This guide explains how to make a CompleXcope application with Open GL and CAVE library. (author)

R&D 100 Awards Demonstrate Clean Energy Legacy - Continuum Magazine |

Science.gov (United States)

Intel to develop an innovative warm-water, liquid-cooled supercomputer that later won an R&D 100 Award. Photo by Dennis Schroeder, NREL R&D 100 Awards Demonstrate Clean Energy Legacy NREL has won 57 R&D 100 Awards since 1982, many of which led directly to industry successes today. R&D 100

Women in Energy: Rinku Gupta - Argonne Today

Science.gov (United States)

-performance clusters and supercomputers. What is the best part of your job? The best part is working with Argonne Today Argonne Today Mission People Work/Life Connections Focal Point Women in Energy: Rinku Gupta Home People Women in Energy: Rinku Gupta Women in Energy: Rinku Gupta Apr 1, 2016 | Posted by Argonne

Kids at CERN Grids for Kids programme leads to advanced computing knowledge.

CERN Multimedia

2008-01-01

Children as young as 10 are learning computing skills, such as middleware, parallel processing and supercomputing, at CERN, the European Organisation for Nuclear Research, last week. The initiative for 10 to 12 years olds is part of the Grids for Kids programme, which aims to introduce Grid computing as a tool for research.

NASA Center for Climate Simulation (NCCS) Presentation

Science.gov (United States)

Webster, William P.

2012-01-01

The NASA Center for Climate Simulation (NCCS) offers integrated supercomputing, visualization, and data interaction technologies to enhance NASA's weather and climate prediction capabilities. It serves hundreds of users at NASA Goddard Space Flight Center, as well as other NASA centers, laboratories, and universities across the US. Over the past year, NCCS has continued expanding its data-centric computing environment to meet the increasingly data-intensive challenges of climate science. We doubled our Discover supercomputer's peak performance to more than 800 teraflops by adding 7,680 Intel Xeon Sandy Bridge processor-cores and most recently 240 Intel Xeon Phi Many Integrated Core (MIG) co-processors. A supercomputing-class analysis system named Dali gives users rapid access to their data on Discover and high-performance software including the Ultra-scale Visualization Climate Data Analysis Tools (UV-CDAT), with interfaces from user desktops and a 17- by 6-foot visualization wall. NCCS also is exploring highly efficient climate data services and management with a new MapReduce/Hadoop cluster while augmenting its data distribution to the science community. Using NCCS resources, NASA completed its modeling contributions to the Intergovernmental Panel on Climate Change (IPCG) Fifth Assessment Report this summer as part of the ongoing Coupled Modellntercomparison Project Phase 5 (CMIP5). Ensembles of simulations run on Discover reached back to the year 1000 to test model accuracy and projected climate change through the year 2300 based on four different scenarios of greenhouse gases, aerosols, and land use. The data resulting from several thousand IPCC/CMIP5 simulations, as well as a variety of other simulation, reanalysis, and observationdatasets, are available to scientists and decision makers through an enhanced NCCS Earth System Grid Federation Gateway. Worldwide downloads have totaled over 110 terabytes of data.

Parallel-vector algorithms for particle simulations on shared-memory multiprocessors

International Nuclear Information System (INIS)

Nishiura, Daisuke; Sakaguchi, Hide

2011-01-01

Over the last few decades, the computational demands of massive particle-based simulations for both scientific and industrial purposes have been continuously increasing. Hence, considerable efforts are being made to develop parallel computing techniques on various platforms. In such simulations, particles freely move within a given space, and so on a distributed-memory system, load balancing, i.e., assigning an equal number of particles to each processor, is not guaranteed. However, shared-memory systems achieve better load balancing for particle models, but suffer from the intrinsic drawback of memory access competition, particularly during (1) paring of contact candidates from among neighboring particles and (2) force summation for each particle. Here, novel algorithms are proposed to overcome these two problems. For the first problem, the key is a pre-conditioning process during which particle labels are sorted by a cell label in the domain to which the particles belong. Then, a list of contact candidates is constructed by pairing the sorted particle labels. For the latter problem, a table comprising the list indexes of the contact candidate pairs is created and used to sum the contact forces acting on each particle for all contacts according to Newton's third law. With just these methods, memory access competition is avoided without additional redundant procedures. The parallel efficiency and compatibility of these two algorithms were evaluated in discrete element method (DEM) simulations on four types of shared-memory parallel computers: a multicore multiprocessor computer, scalar supercomputer, vector supercomputer, and graphics processing unit. The computational efficiency of a DEM code was found to be drastically improved with our algorithms on all but the scalar supercomputer. Thus, the developed parallel algorithms are useful on shared-memory parallel computers with sufficient memory bandwidth.

Multibillion-atom Molecular Dynamics Simulations of Plasticity, Spall, and Ejecta

Science.gov (United States)

Germann, Timothy C.

2007-06-01

Modern supercomputing platforms, such as the IBM BlueGene/L at Lawrence Livermore National Laboratory and the Roadrunner hybrid supercomputer being built at Los Alamos National Laboratory, are enabling large-scale classical molecular dynamics simulations of phenomena that were unthinkable just a few years ago. Using either the embedded atom method (EAM) description of simple (close-packed) metals, or modified EAM (MEAM) models of more complex solids and alloys with mixed covalent and metallic character, simulations containing billions to trillions of atoms are now practical, reaching volumes in excess of a cubic micron. In order to obtain any new physical insights, however, it is equally important that the analysis of such systems be tractable. This is in fact possible, in large part due to our highly efficient parallel visualization code, which enables the rendering of atomic spheres, Eulerian cells, and other geometric objects in a matter of minutes, even for tens of thousands of processors and billions of atoms. After briefly describing the BlueGene/L and Roadrunner architectures, and the code optimization strategies that were employed, results obtained thus far on BlueGene/L will be reviewed, including: (1) shock compression and release of a defective EAM Cu sample, illustrating the plastic deformation accompanying void collapse as well as the subsequent void growth and linkup upon release; (2) solid-solid martensitic phase transition in shock-compressed MEAM Ga; and (3) Rayleigh-Taylor fluid instability modeled using large-scale direct simulation Monte Carlo (DSMC) simulations. I will also describe our initial experiences utilizing Cell Broadband Engine processors (developed for the Sony PlayStation 3), and planned simulation studies of ejecta and spall failure in polycrystalline metals that will be carried out when the full Petaflop Opteron/Cell Roadrunner supercomputer is assembled in mid-2008.

Federated data storage system prototype for LHC experiments and data intensive science

Science.gov (United States)

Kiryanov, A.; Klimentov, A.; Krasnopevtsev, D.; Ryabinkin, E.; Zarochentsev, A.

2017-10-01

Rapid increase of data volume from the experiments running at the Large Hadron Collider (LHC) prompted physics computing community to evaluate new data handling and processing solutions. Russian grid sites and universities’ clusters scattered over a large area aim at the task of uniting their resources for future productive work, at the same time giving an opportunity to support large physics collaborations. In our project we address the fundamental problem of designing a computing architecture to integrate distributed storage resources for LHC experiments and other data-intensive science applications and to provide access to data from heterogeneous computing facilities. Studies include development and implementation of federated data storage prototype for Worldwide LHC Computing Grid (WLCG) centres of different levels and University clusters within one National Cloud. The prototype is based on computing resources located in Moscow, Dubna, Saint Petersburg, Gatchina and Geneva. This project intends to implement a federated distributed storage for all kind of operations such as read/write/transfer and access via WAN from Grid centres, university clusters, supercomputers, academic and commercial clouds. The efficiency and performance of the system are demonstrated using synthetic and experiment-specific tests including real data processing and analysis workflows from ATLAS and ALICE experiments, as well as compute-intensive bioinformatics applications (PALEOMIX) running on supercomputers. We present topology and architecture of the designed system, report performance and statistics for different access patterns and show how federated data storage can be used efficiently by physicists and biologists. We also describe how sharing data on a widely distributed storage system can lead to a new computing model and reformations of computing style, for instance how bioinformatics program running on supercomputers can read/write data from the federated storage.

A new synthetic approach to the science of complexity: the MISSION project at NIFS

International Nuclear Information System (INIS)

Tetsuya Sato

1999-01-01

The present day stage of computer simulation has entered into the third phase. The age of computer simulation in plasma physics dawned in the late 1950's when Oscar Buneman and John Dawson developed the sheet particle model. In 1960's and the early part of 1970's, one devoted oneself to refine the particle model and to develop, more practically, fluid magnetic hydrodynamic models so that the feasibility of the computer simulation methodology could be tested. This age can be called 'dawning of computer simulation'. The so-called supercomputer appeared in the late 1970's and the computer simulation entered the second phase where individual nonlinear phenomena have become possible to be attacked. At present when a supercomputer with the ability of higher than 100 GFlops speed and lager than 10GBytes common memory is available, almost any individual nonlinear phenomenon, whatever it may look complex, can be solved. This age may well be called 'the age of nonlinear solver'. However, as far as the authors are satisfied with using a supercomputer for simply solving an individual nonlinear problem, the computer simulation plays only a passive role in science and would never cause a catastrophic transition to it. Then, the modern science of the 20th century based on reductionism would continue in the coming 21st century, thus the 21st century's science would stay boring and tedious. It must be the computer simulation that can refresh this boring state. At NIFS an extensive effort has been made to establish a new paradigm of science in the 21st century by developing a new synthetic methodology of computer simulation, which the authors call the MISSION Project. The authors present this MISSION Project and propose a working hypothesis of the science of complexity in this talk

Forschungszentrum Juelich. Annual report 2008

International Nuclear Information System (INIS)

Frick, Frank; Roegener, Wiebke

2009-07-01

The following topics are dealt with: The precise lattice QCD mass calculation of protons and neutrons by means of the JUGENE supercomputer, the early diagnosis of morbus Alzheimer, the fabrication of vertebra-column implants consisting of porus titanium, software for the improvement of the spatial resolution in electron microscopy by means of aberration corrections. (HSI)

Vector and parallel processors in computational science

International Nuclear Information System (INIS)

Duff, I.S.; Reid, J.K.

1985-01-01

This book presents the papers given at a conference which reviewed the new developments in parallel and vector processing. Topics considered at the conference included hardware (array processors, supercomputers), programming languages, software aids, numerical methods (e.g., Monte Carlo algorithms, iterative methods, finite elements, optimization), and applications (e.g., neutron transport theory, meteorology, image processing)

Lessons learned from the K computer project—from the K computer to Exascale

International Nuclear Information System (INIS)

Oyanagi, Yoshio

2014-01-01

The history of the supercomputers in Japan and the U.S. is briefly summarized and the difference between the two is discussed. The development of the K Computer project in Japan is described as compared to other PetaFlops projects. The difficulties to be solved in the Exascale computer project now being developed are discussed

Forschungszentrum Juelich. Annual report 2008; Forschungszentrum Juelich. Jahresbericht 2008

Energy Technology Data Exchange (ETDEWEB)

Frick, Frank; Roegener, Wiebke

2009-07-15

The following topics are dealt with: The precise lattice QCD mass calculation of protons and neutrons by means of the JUGENE supercomputer, the early diagnosis of morbus Alzheimer, the fabrication of vertebra-column implants consisting of porus titanium, software for the improvement of the spatial resolution in electron microscopy by means of aberration corrections. (HSI)

New frontiers in nuclear structure studies

International Nuclear Information System (INIS)

Zwarts, D.; Walet, N.R.; Wolters, A.A.; Glaudemans, P.W.M.; VandeGraff, R.J.

1985-01-01

The need to go to larger model spaces for more detailed studies of the atomic nucleus has led to the introduction of the supercomputer to nuclear physics. In this report a brief survey of the nuclear shell model is presented and the performance of some of the relevant programs on different computer systems is compared

Visualization system for grid environment in the nuclear field

International Nuclear Information System (INIS)

Suzuki, Yoshio; Matsumoto, Nobuko; Idomura, Yasuhiro; Tani, Masayuki

2006-01-01

An innovative scientific visualization system is needed to integratedly visualize large amount of data which are distributedly generated in remote locations as a result of a large-scale numerical simulation using a grid environment. One of the important functions in such a visualization system is a parallel visualization which enables to visualize data using multiple CPUs of a supercomputer. The other is a distributed visualization which enables to execute visualization processes using a local client computer and remote computers. We have developed a toolkit including these functions in cooperation with the commercial visualization software AVS/Express, called Parallel Support Toolkit (PST). PST can execute visualization processes with three kinds of parallelism (data parallelism, task parallelism and pipeline parallelism) using local and remote computers. We have evaluated PST for large amount of data generated by a nuclear fusion simulation. Here, two supercomputers Altix3700Bx2 and Prism installed in JAEA are used. From the evaluation, it can be seen that PST has a potential to efficiently visualize large amount of data in a grid environment. (author)

Multi-threaded ATLAS simulation on Intel Knights Landing processors

CERN Document Server

AUTHOR|(INSPIRE)INSPIRE-00014247; The ATLAS collaboration; Calafiura, Paolo; Leggett, Charles; Tsulaia, Vakhtang; Dotti, Andrea

2017-01-01

The Knights Landing (KNL) release of the Intel Many Integrated Core (MIC) Xeon Phi line of processors is a potential game changer for HEP computing. With 72 cores and deep vector registers, the KNL cards promise significant performance benefits for highly-parallel, compute-heavy applications. Cori, the newest supercomputer at the National Energy Research Scientific Computing Center (NERSC), was delivered to its users in two phases with the first phase online at the end of 2015 and the second phase now online at the end of 2016. Cori Phase 2 is based on the KNL architecture and contains over 9000 compute nodes with 96GB DDR4 memory. ATLAS simulation with the multithreaded Athena Framework (AthenaMT) is a good potential use-case for the KNL architecture and supercomputers like Cori. ATLAS simulation jobs have a high ratio of CPU computation to disk I/O and have been shown to scale well in multi-threading and across many nodes. In this paper we will give an overview of the ATLAS simulation application with detai...

Multi-threaded ATLAS Simulation on Intel Knights Landing Processors

CERN Document Server

Farrell, Steven; The ATLAS collaboration; Calafiura, Paolo; Leggett, Charles

2016-01-01

The Knights Landing (KNL) release of the Intel Many Integrated Core (MIC) Xeon Phi line of processors is a potential game changer for HEP computing. With 72 cores and deep vector registers, the KNL cards promise significant performance benefits for highly-parallel, compute-heavy applications. Cori, the newest supercomputer at the National Energy Research Scientific Computing Center (NERSC), will be delivered to its users in two phases with the first phase online now and the second phase expected in mid-2016. Cori Phase 2 will be based on the KNL architecture and will contain over 9000 compute nodes with 96GB DDR4 memory. ATLAS simulation with the multithreaded Athena Framework (AthenaMT) is a great use-case for the KNL architecture and supercomputers like Cori. Simulation jobs have a high ratio of CPU computation to disk I/O and have been shown to scale well in multi-threading and across many nodes. In this presentation we will give an overview of the ATLAS simulation application with details on its multi-thr...

The NEST Dry-Run Mode: Efficient Dynamic Analysis of Neuronal Network Simulation Code

Directory of Open Access Journals (Sweden)

Susanne Kunkel

2017-06-01

Full Text Available NEST is a simulator for spiking neuronal networks that commits to a general purpose approach: It allows for high flexibility in the design of network models, and its applications range from small-scale simulations on laptops to brain-scale simulations on supercomputers. Hence, developers need to test their code for various use cases and ensure that changes to code do not impair scalability. However, running a full set of benchmarks on a supercomputer takes up precious compute-time resources and can entail long queuing times. Here, we present the NEST dry-run mode, which enables comprehensive dynamic code analysis without requiring access to high-performance computing facilities. A dry-run simulation is carried out by a single process, which performs all simulation steps except communication as if it was part of a parallel environment with many processes. We show that measurements of memory usage and runtime of neuronal network simulations closely match the corresponding dry-run data. Furthermore, we demonstrate the successful application of the dry-run mode in the areas of profiling and performance modeling.

Integration of the Chinese HPC Grid in ATLAS Distributed Computing

Science.gov (United States)

Filipčič, A.; ATLAS Collaboration

2017-10-01

Fifteen Chinese High-Performance Computing sites, many of them on the TOP500 list of most powerful supercomputers, are integrated into a common infrastructure providing coherent access to a user through an interface based on a RESTful interface called SCEAPI. These resources have been integrated into the ATLAS Grid production system using a bridge between ATLAS and SCEAPI which translates the authorization and job submission protocols between the two environments. The ARC Computing Element (ARC-CE) forms the bridge using an extended batch system interface to allow job submission to SCEAPI. The ARC-CE was setup at the Institute for High Energy Physics, Beijing, in order to be as close as possible to the SCEAPI front-end interface at the Computing Network Information Center, also in Beijing. This paper describes the technical details of the integration between ARC-CE and SCEAPI and presents results so far with two supercomputer centers, Tianhe-IA and ERA. These two centers have been the pilots for ATLAS Monte Carlo Simulation in SCEAPI and have been providing CPU power since fall 2015.

Multi-level programming paradigm for extreme computing

International Nuclear Information System (INIS)

Petiton, S.; Sato, M.; Emad, N.; Calvin, C.; Tsuji, M.; Dandouna, M.

2013-01-01

In order to propose a framework and programming paradigms for post peta-scale computing, on the road to exa-scale computing and beyond, we introduced new languages, associated with a hierarchical multi-level programming paradigm, allowing scientific end-users and developers to program highly hierarchical architectures designed for extreme computing. In this paper, we explain the interest of such hierarchical multi-level programming paradigm for extreme computing and its well adaptation to several large computational science applications, such as for linear algebra solvers used for reactor core physic. We describe the YML language and framework allowing describing graphs of parallel components, which may be developed using PGAS-like language such as XMP, scheduled and computed on supercomputers. Then, we propose experimentations on supercomputers (such as the 'K' and 'Hooper' ones) of the hybrid method MERAM (Multiple Explicitly Restarted Arnoldi Method) as a case study for iterative methods manipulating sparse matrices, and the block Gauss-Jordan method as a case study for direct method manipulating dense matrices. We conclude proposing evolutions for this programming paradigm. (authors)

Performance evaluation of scientific programs on advanced architecture computers

International Nuclear Information System (INIS)

Walker, D.W.; Messina, P.; Baille, C.F.

1988-01-01

Recently a number of advanced architecture machines have become commercially available. These new machines promise better cost-performance then traditional computers, and some of them have the potential of competing with current supercomputers, such as the Cray X/MP, in terms of maximum performance. This paper describes an on-going project to evaluate a broad range of advanced architecture computers using a number of complete scientific application programs. The computers to be evaluated include distributed- memory machines such as the NCUBE, INTEL and Caltech/JPL hypercubes, and the MEIKO computing surface, shared-memory, bus architecture machines such as the Sequent Balance and the Alliant, very long instruction word machines such as the Multiflow Trace 7/200 computer, traditional supercomputers such as the Cray X.MP and Cray-2, and SIMD machines such as the Connection Machine. Currently 11 application codes from a number of scientific disciplines have been selected, although it is not intended to run all codes on all machines. Results are presented for two of the codes (QCD and missile tracking), and future work is proposed

Optimizing the performance of streaming numerical kernels on the IBM Blue Gene/P PowerPC 450 processor

KAUST Repository

Malas, Tareq Majed Yasin

2012-05-21

Several emerging petascale architectures use energy-efficient processors with vectorized computational units and in-order thread processing. On these architectures the sustained performance of streaming numerical kernels, ubiquitous in the solution of partial differential equations, represents a challenge despite the regularity of memory access. Sophisticated optimization techniques are required to fully utilize the CPU. We propose a new method for constructing streaming numerical kernels using a high-level assembly synthesis and optimization framework. We describe an implementation of this method in Python targeting the IBM® Blue Gene®/P supercomputer\\'s PowerPC® 450 core. This paper details the high-level design, construction, simulation, verification, and analysis of these kernels utilizing a subset of the CPU\\'s instruction set. We demonstrate the effectiveness of our approach by implementing several three-dimensional stencil kernels over a variety of cached memory scenarios and analyzing the mechanically scheduled variants, including a 27-point stencil achieving a 1.7× speedup over the best previously published results. © The Author(s) 2012.

Particle Communication and Domain Neighbor Coupling: Scalable Domain Decomposed Algorithms for Monte Carlo Particle Transport

Energy Technology Data Exchange (ETDEWEB)

O' Brien, M. J.; Brantley, P. S.

2015-01-20

In order to run Monte Carlo particle transport calculations on new supercomputers with hundreds of thousands or millions of processors, care must be taken to implement scalable algorithms. This means that the algorithms must continue to perform well as the processor count increases. In this paper, we examine the scalability of:(1) globally resolving the particle locations on the correct processor, (2) deciding that particle streaming communication has finished, and (3) efficiently coupling neighbor domains together with different replication levels. We have run domain decomposed Monte Carlo particle transport on up to 2²¹ = 2,097,152 MPI processes on the IBM BG/Q Sequoia supercomputer and observed scalable results that agree with our theoretical predictions. These calculations were carefully constructed to have the same amount of work on every processor, i.e. the calculation is already load balanced. We also examine load imbalanced calculations where each domain’s replication level is proportional to its particle workload. In this case we show how to efficiently couple together adjacent domains to maintain within workgroup load balance and minimize memory usage.

Computational complexity of the landscape II-Cosmological considerations

Science.gov (United States)

Denef, Frederik; Douglas, Michael R.; Greene, Brian; Zukowski, Claire

2018-05-01

We propose a new approach for multiverse analysis based on computational complexity, which leads to a new family of "computational" measure factors. By defining a cosmology as a space-time containing a vacuum with specified properties (for example small cosmological constant) together with rules for how time evolution will produce the vacuum, we can associate global time in a multiverse with clock time on a supercomputer which simulates it. We argue for a principle of "limited computational complexity" governing early universe dynamics as simulated by this supercomputer, which translates to a global measure for regulating the infinities of eternal inflation. The rules for time evolution can be thought of as a search algorithm, whose details should be constrained by a stronger principle of "minimal computational complexity". Unlike previously studied global measures, ours avoids standard equilibrium considerations and the well-known problems of Boltzmann Brains and the youngness paradox. We also give various definitions of the computational complexity of a cosmology, and argue that there are only a few natural complexity classes.

Control, data acquisition, data analysis and remote participation in LHD

International Nuclear Information System (INIS)

Nagayama, Y.; Emoto, M.; Nakanishi, H.; Sudo, S.; Imazu, S.; Inagaki, S.; Iwata, C.; Kojima, M.; Nonomura, M.; Ohsuna, M.; Tsuda, K.; Yoshida, M.; Chikaraishi, H.; Funaba, H.; Horiuchi, R.; Ishiguro, S.; Ito, Y.; Kubo, S.; Mase, A.; Mito, T.

2008-01-01

This paper presents the control, data acquisition, data analysis and remote participation facilities of the Large Helical Device (LHD), which is designed to confine the plasma in steady state. In LHD the plasma duration exceeds 3000 s by controlling the plasma position, the density and the ICRF heating. The 'LABCOM' data acquisition system takes both the short-pulse and the steady-state data. A two-layer Mass Storage System with RAIDs and Blu-ray Disk jukeboxes in a storage area network has been developed to increase capacity of storage. The steady-state data can be monitored with a Web browser in real time. A high-level data analysis system with Web interfaces is being developed in order to provide easier usage of LHD data and large FORTRAN codes in a supercomputer. A virtual laboratory system for the Japanese fusion community has been developed with Multi-protocol Label Switching Virtual Private Network Technology. Collaborators at remote sites can join the LHD experiment or use the NIFS supercomputer system as if they were working in the LHD control room

Subsampling-based compression and flow visualization

Energy Technology Data Exchange (ETDEWEB)

Agranovsky, Alexy; Camp, David; Joy, I; Childs, Hank

2016-01-19

As computational capabilities increasingly outpace disk speeds on leading supercomputers, scientists will, in turn, be increasingly unable to save their simulation data at its native resolution. One solution to this problem is to compress these data sets as they are generated and visualize the compressed results afterwards. We explore this approach, specifically subsampling velocity data and the resulting errors for particle advection-based flow visualization. We compare three techniques: random selection of subsamples, selection at regular locations corresponding to multi-resolution reduction, and introduce a novel technique for informed selection of subsamples. Furthermore, we explore an adaptive system which exchanges the subsampling budget over parallel tasks, to ensure that subsampling occurs at the highest rate in the areas that need it most. We perform supercomputing runs to measure the effectiveness of the selection and adaptation techniques. Overall, we find that adaptation is very effective, and, among selection techniques, our informed selection provides the most accurate results, followed by the multi-resolution selection, and with the worst accuracy coming from random subsamples.

Proceedings of the 2. international conference on simulation methods in nuclear engineering

International Nuclear Information System (INIS)

Brais, A.

1986-10-01

The fifty papers presented at this conference cover the field of computerized simulation and mathematical modelling of processes in PWR and CANDU type reactors. Topics include thermalhydraulics of system transients, complex geometries, multi-fluid systems, and general situations; reactor physics simulations; reactor simulators; fuel and fuel channel behaviour; and applications of supercomputers, parallel processing, and microcomputers

Engineering Technology Showcase to feature high-tech products from 18 companies

OpenAIRE

Gilbert, Karen

2007-01-01

The Student Technology Council (STC) at Virginia Tech is sponsoring an Engineering Technology Showcase on Tuesday, March 27. In addition to providing a platform for technology companies to show off their most recent innovations, technology presentations will be offered by Virginia Tech faculty and staff on topics ranging from a virtual greenhouse to the System X supercomputer.

Guide for 3D WARP simulations of hollow electron beam lenses. Practical explanation on basis of Tevatron electron lens test stand

Energy Technology Data Exchange (ETDEWEB)

Moens, Vince [Ecole Polytechnique Federale de Lausanne (EPFL) (Switzerland)

2014-06-08

The purpose of this guide is to help successive students handle WARP. It outlines the installation of WARP on personal computers as well as super-computers and clusters. It furthermore teaches the reader how to handle the WARP environment and run basic scripts. Lastly it outlines how to execute the current Hollow Electron Beam Lens scripts.

Life, the universe and lots of data

CERN Multimedia

Nairn, G

2003-01-01

CERN's latest project, the large hadron collider, will produce data in such quantities when it is switched on in 2007 that not even a battery of supercomputers could handle it all. Instead, Cern plans to create a massive computing grid that will harness the power and storage capacity of tens of thousands of computers around the world (1 page).

Sieving for pseudosquares and pseudocubes in parallel using doubly-focused enumeration and wheel datastructures

OpenAIRE

Sorenson, Jonathan P.

2010-01-01

We extend the known tables of pseudosquares and pseudocubes, discuss the implications of these new data on the conjectured distribution of pseudosquares and pseudocubes, and present the details of the algorithm used to do this work. Our algorithm is based on the space-saving wheel data structure combined with doubly-focused enumeration, run in parallel on a cluster supercomputer.

SERS internship fall 1995 abstracts and research papers

Energy Technology Data Exchange (ETDEWEB)

Davis, Beverly

1996-05-01

This report is a compilation of twenty abstracts and their corresponding full papers of research projects done under the US Department of Energy Science and Engineering Research Semester (SERS) program. Papers cover a broad range of topics, for example, environmental transport, supercomputers, databases, biology. Selected papers were indexed separately for inclusion the the Energy Science and Technology Database.

Sparx PCA Module

Energy Technology Data Exchange (ETDEWEB)

2017-04-25

Sparx, a new environment for Cryo-EM image processing; Cryo-EM, Single particle reconstruction, principal component analysis; Hardware Req.: PC, MAC, Supercomputer, Mainframe, Multiplatform, Workstation. Software Req.: operating system is Unix; Compiler C++; type of files: source code, object library, executable modules, compilation instructions; sample problem input data. Location/transmission: http://sparx-em.org; User manual & paper: http://sparx-em.org;

Implementation and qualification of MCNP 5 through the intercomparison with the benchmark for the calculation of critical systems Godiva and Jezebel

International Nuclear Information System (INIS)

Lara, Rafael G.; Maiorino, Jose R.

2013-01-01

This work aimed at the implementation and qualification of MCNP code in a supercomputer of the Universidade Federal do ABC, so that may be available a next-generation simulation tool for precise calculations of nuclear reactors and systems subject to radiation. The implementation of this tool will have multidisciplinary applications, covering various areas of engineering (nuclear, aerospace, biomedical), radiation physics and others

Holistic Approach to Data Center Energy Efficiency

Energy Technology Data Exchange (ETDEWEB)

Hammond, Steven W [National Renewable Energy Laboratory (NREL), Golden, CO (United States)

2017-09-18

This presentation discusses NREL's Energy System Integrations Facility and NREL's holistic design approach to sustainable data centers that led to the world's most energy-efficient data center. It describes Peregrine, a warm water liquid cooled supercomputer, waste heat reuse in the data center, demonstrated PUE and ERE, and lessons learned during four years of operation.

Implementation and qualification of MCNP 5 through the intercomparison with the benchmark for the calculation of critical systems Godiva and Jezebel; Implementacao e qualificacao do MCNP5 atraves da intercomparacao com o benchmark para o calculo dos sistemas criticos Godiva e Jezebel

Energy Technology Data Exchange (ETDEWEB)

Lara, Rafael G.; Maiorino, Jose R., E-mail: rafael.lara@aluno.ufabc.edu.br, E-mail: joserubens.maiorino@ufabc.edu.br [Universidade Federal do ABC (UFABC), Santo Andre, SP (Brazil). Centro de Engenharia, Modelagem e Ciencias Sociais Aplicadas

2013-07-01

This work aimed at the implementation and qualification of MCNP code in a supercomputer of the Universidade Federal do ABC, so that may be available a next-generation simulation tool for precise calculations of nuclear reactors and systems subject to radiation. The implementation of this tool will have multidisciplinary applications, covering various areas of engineering (nuclear, aerospace, biomedical), radiation physics and others.

Proceedings of the ninth annual conference of the IEEE Engineering in Medicine and Biology Society

International Nuclear Information System (INIS)

Anon.

1987-01-01

This book contains over 100 papers. Some of the titles are: Angular integrations and inter-projections correlation effects in CT reconstruction; Supercomputing environment for biomedical research; Program towards a computational molecular biology; Current problems in molecular biology computing; Signal averaging applied to positron emission tomography; First experimental results from a high spatial resolution PET prototype; and A coherent approach in computer-aided radiotherapy

Computers are stepping stones to improved imaging.

Science.gov (United States)

Freiherr, G

1991-02-01

Never before has the radiology industry embraced the computer with such enthusiasm. Graphics supercomputers as well as UNIX- and RISC-based computing platforms are turning up in every digital imaging modality and especially in systems designed to enhance and transmit images, says author Greg Freiherr on assignment for Computers in Healthcare at the Radiological Society of North America conference in Chicago.

Supercomputing for molecular dynamics simulations handling multi-trillion particles in nanofluidics

CERN Document Server

Heinecke, Alexander; Horsch, Martin; Bungartz, Hans-Joachim

2015-01-01

This work presents modern implementations of relevant molecular dynamics algorithms using ls1 mardyn, a simulation program for engineering applications. The text focuses strictly on HPC-related aspects, covering implementation on HPC architectures, taking Intel Xeon and Intel Xeon Phi clusters as representatives of current platforms. The work describes distributed and shared-memory parallelization on these platforms, including load balancing, with a particular focus on the efficient implementation of the compute kernels. The text also discusses the software-architecture of the resulting code.

Plasma physics on the TI-85 calculator or Down with supercomputers

International Nuclear Information System (INIS)

Sedlacek, Z.

1998-10-01

In the Fourier transformed velocity space the Vlasov plasma oscillations may be interpreted as a wave propagation process corresponding to an imperfectly trapped (leaking) wave. The Landau damped solutions of the Vlasov-Poisson equation then become genuine Eigenmodes corresponding to complex eigenvalues. To illustrate this new interpretation we solve numerically the Fourier transformed Vlasov-Poisson equation, essentially a perturbed advective equation, on the TI-85 pocket graphics calculator. A program is described, based on the method of lines: A finite-difference scheme is utilized to discretize the transformed equation and the resulting set of ordinary differential equations is then solved in time. The user can choose from several possible finite-difference differentiation schemes differing in the total number of points and the number of downwind points. The resulting evolution of the electric field showing the Landau damped plasma oscillations is displayed on the screen of the calculator. In addition, calculation of the eigenvalues of the Fourier transformed Vlasov-Poisson operator is possible. The user can also experiment with the numerical solution of the advective equation which describes free streaming. (author)

The Erasmus Computing Grid – Building a Super-Computer for Free

NARCIS (Netherlands)

T.A. Knoch (Tobias); A. Abuseiris (Anis); R.M. de Graaf (Rob); M. Lesnussa (Michael); F.G. Grosveld (Frank)

2011-01-01

textabstractToday advances in scientific research as well as clinical diagnostics and treatment are inevitably connected with information solutions concerning computation power and information storage. The needs for information technology are enormous and are in many cases the limiting factor for

Integration Of PanDA Workload Management System With Supercomputers for ATLAS

CERN Document Server

Oleynik, Danila; The ATLAS collaboration; De, Kaushik; Wenaus, Torre; Maeno, Tadashi; Barreiro Megino, Fernando Harald; Nilsson, Paul; Guan, Wen; Panitkin, Sergey

2016-01-01

The Large Hadron Collider (LHC), operating at the international CERN Laboratory in Geneva, Switzerland, is leading Big Data driven scientific explorations. Experiments at the LHC explore the fundamental nature of matter and the basic forces that shape our universe, and were recently credited for the discovery of a Higgs boson. ATLAS, one of the largest collaborations ever assembled in the sciences, is at the forefront of research at the LHC. To address an unprecedented multi-petabyte data processing challenge, the ATLAS experiment is relying on a heterogeneous distributed computational infrastructure. The ATLAS experiment uses PanDA (Production ANd Distributed Analysis system) Workload Management System for managing the workflow for all data processing on over 150 data centers. Through PanDA, ATLAS physicists see a single computing facility that enables rapid scientific breakthroughs for the experiment, even though the data centers are physically scattered all over the world. While PanDA currently uses more t...

Supercomputer methods for the solution of fundamental problems of particle physics

International Nuclear Information System (INIS)

Moriarty, K.J.M.; Rebbi, C.

1990-01-01

The authors present motivation and methods for computer investigations in particle theory. They illustrate the computational formulation of quantum chromodynamics and selected application to the calculation of hadronic properties. They discuss possible extensions of the methods developed for particle theory to different areas of applications, such as cosmology and solid-state physics, that share common methods. Because of the commonality of methodology, advances in one area stimulate advances in other ares. They also outline future plans of research

Advances in Supercomputing for the Modeling of Atomic Processes in Plasmas

International Nuclear Information System (INIS)

Ludlow, J. A.; Ballance, C. P.; Loch, S. D.; Lee, T. G.; Pindzola, M. S.; Griffin, D. C.; McLaughlin, B. M.; Colgan, J.

2009-01-01

An overview will be given of recent atomic and molecular collision methods developed to take advantage of modern massively parallel computers. The focus will be on direct solutions of the time-dependent Schroedinger equation for simple systems using large numerical lattices, as found in the time-dependent close-coupling method, and for configuration interaction solutions of the time-independent Schroedinger equation for more complex systems using large numbers of basis functions, as found in the R-matrix with pseudo-states method. Results from these large scale calculations are extremely useful in benchmarking less accurate theoretical methods and experimental data. To take full advantage of future petascale and exascale computing resources, it appears that even finer grain parallelism will be needed.

Connecting the dots, or nuclear data in the age of supercomputing

International Nuclear Information System (INIS)

Bauge, E.; Dupuis, M.; Hilaire, S.; Peru, S.; Koning, A.J.; Rochman, D.; Goriely, S.

2014-01-01

Recent increases in computing power have allowed for much progress to be made in the field of nuclear data. The advances listed below are each significant, but together bring the potential to completely change our perspective on the nuclear data evaluation process. The use of modern nuclear modeling codes like TALYS and the Monte Carlo sampling of its model parameter space, together with a code system developed at NRG Petten, which automates the production of ENDF-6 formatted files, their processing, and their use in nuclear reactor calculations, constitutes the Total Monte Carlo approach, which directly links physical model parameters with calculated integral observables like k_e_f_f. Together with the Backward-Forward Monte Carlo method for weighting samples according their statistical likelihood, the Total Monte Carlo can be applied to complete isotopic chains in a consistent way, to simultaneously evaluate nuclear data and the associated uncertainties in the continuum region. Another improvement is found in the uses of microscopic models for nuclear reaction calculations. For example, making use of QRPA excited states calculated with the Gogny interaction to solve the long standing question of the origin of the ad hoc 'pseudo-states' that are introduced in evaluated nuclear data files to account for the Livermore pulsed sphere experiments. A third advance consists of the recent optimization of the Gogny D1M effective nuclear interaction, including constraints from experimental nuclear masses at the 'beyond the mean field' level. All these advances are only made possible by the availability of vast resources of computing power, and even greater resources will allow connecting them, going continuously from the parameters of the nuclear interaction to reactor calculations. However, such scheme will surely only be usable for applications if a few fine-tuning 'knobs' are introduced in it. The values of these adjusted parameters will have to be calibrated versus differential and integral experimental constraints. (authors)

Programmable lithography engine (ProLE) grid-type supercomputer and its applications

Science.gov (United States)

Petersen, John S.; Maslow, Mark J.; Gerold, David J.; Greenway, Robert T.

2003-06-01

There are many variables that can affect lithographic dependent device yield. Because of this, it is not enough to make optical proximity corrections (OPC) based on the mask type, wavelength, lens, illumination-type and coherence. Resist chemistry and physics along with substrate, exposure, and all post-exposure processing must be considered too. Only a holistic approach to finding imaging solutions will accelerate yield and maximize performance. Since experiments are too costly in both time and money, accomplishing this takes massive amounts of accurate simulation capability. Our solution is to create a workbench that has a set of advanced user applications that utilize best-in-class simulator engines for solving litho-related DFM problems using distributive computing. Our product, ProLE (Programmable Lithography Engine), is an integrated system that combines Petersen Advanced Lithography Inc."s (PAL"s) proprietary applications and cluster management software wrapped around commercial software engines, along with optional commercial hardware and software. It uses the most rigorous lithography simulation engines to solve deep sub-wavelength imaging problems accurately and at speeds that are several orders of magnitude faster than current methods. Specifically, ProLE uses full vector thin-mask aerial image models or when needed, full across source 3D electromagnetic field simulation to make accurate aerial image predictions along with calibrated resist models;. The ProLE workstation from Petersen Advanced Lithography, Inc., is the first commercial product that makes it possible to do these intensive calculations at a fraction of a time previously available thus significantly reducing time to market for advance technology devices. In this work, ProLE is introduced, through model comparison to show why vector imaging and rigorous resist models work better than other less rigorous models, then some applications of that use our distributive computing solution are shown. Topics covered describe why ProLE solutions are needed from an economic and technical aspect, a high level discussion of how the distributive system works, speed benchmarking, and finally, a brief survey of applications including advanced aberrations for lens sensitivity and flare studies, optical-proximity-correction for a bitcell and an application that will allow evaluation of the potential of a design to have systematic failures during fabrication.