DOCSLIB.ORG

NAS Parallel Benchmarks, RNR-94-007

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
NAS Parallel Benchmarks, RNR-94-007 THE NAS PARALLEL BENCHMARKS D Bailey E Barszcz J Barton D Browning R Carter L Dagum R Fato ohi S Fineb erg PFrederickson T Lasinski R Schreib er H Simon V Venkatakrishnan and S Weeratunga RNR Technical Rep ort RNR March Abstract A new set of b enchmarks has b een develop ed for the p erformance eval uation of highly parallel sup ercomputers These b enchmarks consist of ve parallel kernels and three simulated application b enchmarks Together they mimic the computation and data movementcharacteristics of large scale computational uid dynamics CFD applications The principal distinguishing feature of these b enchmarks is their p encil and pap er sp ecicationall details of these b enchmarks are sp ecied only algorithmically In this waymany of the diculties asso ciated with conven tional b enchmarking approaches on highly parallel systems are avoided This pap er is an up dated version of a RNR technical rep ort originally issued January Bailey Barszcz Barton Carter and Lasinski are with NASA Ames ResearchCenter Dagum Fato ohi Fineb erg Simon and Weer atunga are with Computer Sciences Corp work supp orted through NASA Contract NAS Schreib er is with RIACS work funded via NASA co op erative agreement NCC Browning is now in the Netherlands work originally done with CSC Frederickson is now with Cray Research Inc in Los Alamos NM work originally done with RIACS Venkatakrish nan is now with ICASE in Hampton VAwork originally done with CSC Address for all authors except Browning Frederickson and Venkatakrish nan NASA Ames Research Center Mail Stop TA Moett Field CA GENERAL REMARKS D BaileyDBrowning R Carter S Fineb erg and H Simon Intro duction The Numerical Aero dynamic Simulation NAS Program which is based at NASA Ames Research Center is a large scale eort to advance the state of computational aero dynamics Sp ecically the NAS organization aims to provide the Nations aerospace researchanddevelopment communitybythe year a highp erformance op erational computing system capable of sim ulating an entire aerospace vehicle system within a computing time of one to several hours ref p The successful solution of this grand challenge problem will require the development of computer systems that can p erform the required complex scientic computations at a sustained rate nearly one thousand times greater than current generation sup ercomputers can now achieve The architecture of computer systems able to achieve this level of p erformance will likely b e dissimilar to the shared memory multipro cessing sup ercomputers of to day While no consensus yet exists on what the design willbeitislikely that the system will consist of at least pro cessors computing in parallel Highly parallel systems with computing p ower roughly equivalent to tra ditional shared memory multipro cessors exist to day Unfortunately the p er formance evaluation of these systems on comparable typ es of scientic com putations is very dicult for several reasons Few relevant data are available for the p erformance of algorithms of interest to the computational aerophysics community on many currently available parallel systems Benchmarking and p erformance evaluation of such systems has not kept pace with advances in hardware software and algorithms In particular there is as yet no gener ally accepted b enchmark program or even a b enchmark strategy for these systems The p opular kernel b enchmarks that have b een used for traditional vec tor sup ercomputers such as the Livermore Lo ops the LINPACK b ench mark and the original NAS Kernels are clearly inappropriate for Computer Sciences Corp oration El Segundo California This work is supp orted through NASA Contract NAS the p erformance evaluation of highly parallel machines First of all the tuning restrictions of these b enchmarks rule out many widely used parallel extensions More imp ortantly the computation and memory requirements of these programs do not do justice to the vastly increased capabilities of the new parallel machines particularly those systems that will b e available by the mids On the other hand a full scale scientic application is similarly unsuitable First of all p orting a large program to a new parallel computer architecture requires a ma jor eort and it is usually hard to justify a ma jor research task simply to obtain a b enchmark numb er For that reason we b elievethatthe otherwise very successful PERFECT Club b enchmark is not suitable for highly parallel systems This is demonstrated byvery sparse p erformance results for parallel machines in the recent rep orts Alternatively an application b enchmark could assume the availabilityof automatic software to ols for transforming dustydeck source into ecient parallel co de on a variety of systems However such to ols do not exist to day and many scientists doubt that they will ever exist across a wide range of architectures Some other considerations for the development of a meaningful b ench mark for a highly parallel sup ercomputer are the following Advanced parallel systems frequently require new algorithmic and soft ware approaches and these new metho ds are often quite dierent from the conventional metho ds implemented in source co de for a sequential or vector machine Benchmarks must b e generic and should not favor any particular parallel architecture This requirement precludes the usage of any architecturesp ecic co de such as message passing co de The correctness of results and p erformance gures must b e easily veri able This requirement implies that b oth input and output data sets must b e kept very small It also implies that the nature of the compu tation and the exp ected results must b e sp ecied in great detail The memory size and run time requirements must b e easily adjustable to accommo date new systems with increased p ower The b enchmark must b e readily distributable In our view the only b enchmarking approach that satises all of these constraints is a pap er and p encil b enchmark The idea is to sp ecify a set of problems only algorithmicallyEven the input data must b e sp ecied only on pap er Naturally the problem has to b e sp ecied in sucient detail that a unique solution exists and the required output has to b e brief yet detailed enough to certify that the problem has b een solved correctly The p erson or p ersons implementing the b enchmarks on a given system are exp ected to solve the various problems in the most appropriate way for the sp ecic system The choice of data structures algorithms pro cessor allo cation and memory usage are all to the extentallowed by the sp ecication left op en to the discretion of the implementer Some extension of FortranorCisrequired and reasonable limits are placed on the usage of assembly co de and the like but otherwise programmers are free to utilize language constructs that give the b est p erformance p ossible on the particular system b eing studied To this end wehave devised a numb er of relatively simple kernels which are sp ecied completely in chapter of this do cument However kernels alone are insucient to completely assess the p erformance p otential of a parallel machine on real scientic applications The chief dicultyisthat a certain data structure maybevery ecient on a certain system for one of the isolated kernels and yet this data structure would b e inappropriate if incorp orated into a larger application In other words the p erformance of a real computational uid dynamics CFD application on a parallel system is critically dep endent on data motion b etween computational kernels Thus we consider the complete repro duction of this data movementtobeofcritical imp ortance in a b enchmark Our b enchmark set therefore consists of two ma jor comp onents ve par allel kernel b enchmarks and three simulated application b enchmarks The simulated application b enchmarks combine several computations in a man ner that resembles the actual order of execution in certain imp ortantCFD application co des This is discussed in more detail in chapter We feel that this b enchmark set successfully addresses many of the prob lems asso ciated with b enchmarking parallel machines Although wedonot claim that this set is typical of all scientic computing it is based on the key comp onents of several large aeroscience applications used on sup ercomput ers by scientists at NASA Ames ResearchCenter These b enchmarks will b e used by the Numerical Aero dynamic Simulation NAS Program to evaluate the p erformance of parallel computers Benchmark Rules Denitions In the following the term pro cessor is dened as a hardware unit capable of executing b oth oating p oint addition and oating p ointmultiplication in structions The lo cal memory of a pro cessor refers to randomly accessible memory that can b e accessed by that pro cessor in less than one microsecond The term main memory refers to the combined lo cal memory of all pro ces sors This includes any memory shared by all pro cessors that can b e accessed by each pro cessor in less than one microsecond The term mass storage refers to nonvolatile randomly accessible storage media that can b e accessed by at least one pro cessor within forty milliseconds A pro cessing no de is dened as a hardware unit consisting of one or more pro cessors plus their lo cal memory which is logically a single unit on the network that connects the pro cessors The term computational no des refers to those pro cessing no des pri marily devoted to highsp eed oating p oint computation The term ser vice no des refers to those pro cessing no des primarily
Load more

Recommended publications
  • Efficient Multicore Implementation of NAS Benchmarks with Fastflow
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Electronic Thesis and Dissertation Archive - Università di Pisa UNIVERSITY OF PISA AND SCUOLA SUPERIORE SANT'ANNA Master Degree in Computer S ien e an! Net"or#ing Master T$esis Efficient multicore implementation of NAS benchmarks with FastFlow Candi!ate Supervisor Dean De Leo Pro&. Mar o Dane(utto A a!emi Year )*++,)*+) Alla mia famigghia … and to all of the people who have supported me Index 1 Introduction 1 2 Context 4 2.1 NAS Parallel Benchmarks . .4 2.2 FastFlow . .7 2.3 C++ vs Fortran 77 . 10 3 Methodology 12 3.1 Minimising the memory faults . 13 3.2 Optimisation . 17 3.3 Parallel algorithm . 24 4 NPB Kernels 31 4.1 Kernel EP . 32 4.2 Kernel CG . 38 4.3 Kernel MG . 46 5 Experiments 60 5.1 Kernel EP . 62 5.2 Kernel CG . 64 5.3 Kernel MG . 69 6 Conclusions 74 A Source code 77 Chapter 1 Introduction The thesis describes an efficient implementation of a subset of the NAS Parallel Benchmarks (NPB) for the multicore architecture with the FastFlow framework. The NPB is a specification of numeric benchmarks to compare different environ- ments and implementations. FastFlow is a framework, targeted to shared memory systems, supporting the development of parallel algorithms based on the structured parallel programming. Starting from the NPB specification, the thesis selects a subset of the NPB algorithms and discerns an efficient implementation for both the sequential and parallel algorithms, through FastFlow.
    [Show full text]
  • Performance Assessment of Parallel Techniques
    Data Mining VIII: Data, Text and Web Mining and their Business Applications 83 Performance assessment of parallel techniques T. Grytsenko & A. Peratta Wessex Institute of Technology, UK Abstract The goal of this work is to evaluate and compare the computational performance of the most common parallel libraries such as Message Passing Interface (MPI), High Performance Fortran (HPF), OpenMP and DVM for further implementations. Evaluation is based on NAS Parallel benchmark suite (NPB) which includes simulated applications BT, SP, LU and kernel benchmarks FT, CG and MG. A brief introduction of the four parallel techniques under study: MPI, HPF, OpenMP and DVM, as well as their models is provided together with benchmarks used and the test results. Finally, corresponding recommendations are given for the different approaches depending on the number of processors. Keywords: MPI, HPL, DVM, OpenMP, parallel programming, performance, parallel calculations. 1 Introduction This section provides a brief introduction of the four parallel techniques under study: MPI, HPF, OpenMP and DVM, as well as their models. 1.1 The Message-Passing Information programming model (MPI) Programming models are generally categorised according to the way in which how memory is used. In the shared memory model each process accesses a shared address space, while in the message passing model an application runs as a collection of autonomous processes, each with its own local memory. In the message passing model processes communicate with other processes by sending and receiving messages (see Figure 1). When data is passed in a message, the sending and receiving processes must work to transfer the data from the local memory of one to the local memory of the other.
    [Show full text]
  • The Openmp Implementation of NAS Parallel Benchmarks and Its Performance
    The OpenMP Implementation of NAS Parallel Benchmarks and Its Performance H. Jin*, M. Frumkin* and J. Yan* *NASA Ames Research Center, Moffett Field, CA 94035-1000 Abstract As the new ccNUMA architecture became popular in recent years, parallel programming with compiler directives on these machines has evolved to accommodate new needs. In this study, we examine the effectiveness of OpenMP directives for parallelizing the NAS Parallel Benchmarks. Implementation details will be discussed and performance will be compared with the MPI implementation. We have demonstrated that OpenMP can achieve very good results for parallelization on a shared memory system, but effective use of memory and cache is very important. 1. Introduction Over the past decade, high performance computers based on commodity microprocessors have been introduced in rapid succession from many vendors. These machines could be classified into two major categories: distributed memory (DMP) and shared memory (SMP) multiprocessing systems. While both categories consist of physically distributed memories, the programming model on an SMP presents a globally shared address space. SMP's further include hardware supported cache coherence protocols across the processors. The introduction of ccNUMA (cache coherent Non-Uniform Memory Access) architecture, such as SGI Origin 2000, has perspective to scale up to thousands of processors. While all these architectures support some form of message passing (e.g. MPI, the de facto standard today), the SMP's further support program annotation standards, or directives, that allow the user to supply information to the compiler to assist in the parallelization of the code. Currently, there are two accepted standards for annotating programs for parallel executions: HPF and OpenMP.
    [Show full text]
  • Efficient NAS Benchmark Kernels with C++ Parallel Programming
    Efficient NAS Benchmark Kernels with C++ Parallel Programming Dalvan Griebler, Junior Loff, Luiz G. Fernandes Gabriele Mencagli, Marco Danelutto Pontifical Catholic University of Rio Grande do Sul (PUCRS), University of Pisa (UNIPI), Parallel Application Modeling Group (GMAP), Computer Science Department, Porto Alegre – Brazil Pisa – Italy Email: dalvan.griebler,junior.loff @acad.pucrs.br, [email protected] Email: mencagli,marcod @di.unipi.it { } { } Abstract—Benchmarking is a way to study the performance of Our goal is to describe, in a comprehensive way, the new architectures and parallel programming frameworks. Well- porting of the five kernels from the NAS benchmark suite established benchmark suites such as the NAS Parallel Bench- to C++ code as well as the kernels’ parallelization with Intel marks (NPB) comprise legacy codes that still lack portability to C++ language. As a consequence, a set of high-level and easy-to- TBB [8], OpenMP (translation only) [9], and FastFlow [10]. use C++ parallel programming frameworks cannot be tested in After, we aim at evaluating and comparing the performance NPB. Our goal is to describe a C++ porting of the NPB kernels of our implementations with the C++ parallel programming and to analyze the performance achieved by different parallel frameworks. Besides the implementation issues, this work has implementations written using the Intel TBB, OpenMP and a valuable impact regarding research for the following reasons FastFlow frameworks for Multi-Cores. The experiments show an efficient code porting from Fortran to C++ and an efficient that also represent our major contributions: parallelization on average. according to Sec. II, NPB has been ported in different • Index Terms—Parallel Programming, NAS Benchmark, Perfor- languages.
    [Show full text]
  • The Nas Parallel Benchmarks
    THE NAS PARALLEL BENCHMARKS D. Bailey, E. Barszcz, J. Barton, D. Browning, R. Carter, L. Dagum, R. Fatoohi, S. Fineberg, P. Frederickson, T. Lasinski, R. Schreiber, H. Simon, V. Venkatakrishnan and S. Weeratunga Abstract A new set of benchmarks has been developed for the performance eval- uation of highly parallel supercomputers. These benchmarks consist of five parallel kernels and three simulated application benchmarks. Together they mimic the computation and data movement characteristics of large scale computational fluid dynamics (CFD) applications. The principal distinguishing feature of these benchmarks is their “pencil and paper” specification—all details of these benchmarks are specified only algorithmically. In this way many of the difficulties associated with conven- tional benchmarking approaches on highly parallel systems are avoided. This paper is an updated version of a RNR technical report originally issued January 1991. Bailey, Barszcz, Barton, Carter, and Lasinski are with NASA Ames Research Center. Dagum, Fatoohi, Fineberg, Simon and Weer- atunga are with Computer Sciences Corp. (work supported through NASA Contract NAS 2-12961). Schreiber is with RIACS (work funded via NASA cooperative agreement NCC 2-387). Browning is now in the Netherlands (work originally done with CSC). Frederickson is now with Cray Research, Inc. in Los Alamos, NM (work originally done with RIACS). Venkatakrish- nan is now with ICASE in Hampton, VA (work originally done with CSC). Address for all authors except Browning, Frederickson and Venkatakrish- nan: NASA Ames Research Center, Mail Stop T27A-1, Moffett Field, CA 94035-1000. 1 1 GENERAL REMARKS D. Bailey, D. Browning,∗ R. Carter, S. Fineberg,∗ and H. Simon∗ 1.1 Introduction The Numerical Aerodynamic Simulation (NAS) Program, which is based at NASA Ames Research Center, is a large scale effort to advance the state of computational aerodynamics.
    [Show full text]
  • The NAS Parallel Benchmarks
    TITLE: The NAS Parallel Benchmarks AUTHOR: David H Bailey1 ACRONYMS: NAS, NPB DEFINITION: The NAS Parallel Benchmarks (NPB) are a suite of parallel computer per- formance benchmarks. They were originally developed at the NASA Ames Re- search Center in 1991 to assess high-end parallel supercomputers [?]. Although they are no longer used as widely as they once were for comparing high-end sys- tem performance, they continue to be studied and analyzed a great deal in the high-performance computing community. The acronym \NAS" originally stood for the Numerical Aeronautical Simulation Program at NASA Ames. The name of this organization was subsequently changed to the Numerical Aerospace Sim- ulation Program, and more recently to the NASA Advanced Supercomputing Center, although the acronym remains \NAS." The developers of the original NPB suite were David H. Bailey, Eric Barszcz, John Barton, David Browning, Russell Carter, LeoDagum, Rod Fatoohi, Samuel Fineberg, Paul Frederickson, Thomas Lasinski, Rob Schreiber, Horst Simon, V. Venkatakrishnan and Sisira Weeratunga. DISCUSSION: The original NAS Parallel Benchmarks consisted of eight individual bench- mark problems, each of which focused on some aspect of scientific computing. The principal focus was in computational aerophysics, although most of these benchmarks have much broader relevance, since in a much larger sense they are typical of many real-world scientific computing applications. The NPB suite grew out of the need for a more rational procedure to select new supercomputers for acquisition by NASA. The emergence of commercially available highly parallel computer systems in the late 1980s offered an attrac- tive alternative to parallel vector supercomputers that had been the mainstay of high-end scientific computing.
    [Show full text]
  • The Openmp Implementation of NAS Parallel Benchmarks and Its Performance
    The OpenMP Implementation of NAS Parallel Benchmarks and Its Performance ∗ ∗ H. Jin , M. Frumkin and J. Yan NAS Technical Report NAS-99-011 October 1999 {hjin,frumkin,yan}@nas.nasa.gov NAS System Division NASA Ames Research Center Mail Stop T27A-2 Moffett Field, CA 94035-1000 Abstract As the new ccNUMA architecture became popular in recent years, parallel programming with compiler directives on these machines has evolved to accommodate new needs. In this study, we examine the effectiveness of OpenMP directives for parallelizing the NAS Parallel Benchmarks. Implementation details will be discussed and performance will be compared with the MPI implementation. We have demonstrated that OpenMP can achieve very good results for parallelization on a shared memory system, but effective use of memory and cache is very important. Keywords: NAS Parallel Benchmarks, OpenMP program, shared memory system, parallel program performance. ∗ MRJ Technology Solutions, NASA Contract NAS2-14303, Moffett Field, CA 94035-1000 - 2 - The OpenMP Implementation of NAS Parallel Benchmarks and Its Performance ∗ ∗ H. Jin , M. Frumkin and J. Yan NAS Division, NASA Ames Research Center, Moffett Field, CA 94035-1000 Abstract As the new ccNUMA architecture became popular in recent years, parallel programming with compiler directives on these machines has evolved to accommodate new needs. In this study, we examine the effectiveness of OpenMP directives for parallelizing the NAS Parallel Benchmarks. Implementation details will be discussed and performance will be compared with the MPI implementation. We have demonstrated that OpenMP can achieve very good results for parallelization on a shared memory system, but effective use of memory and cache is very important.
    [Show full text]
  • Performance and Scalability of the NAS Parallel Benchmarks in Java
    z Performance and Scalability of the NAS Parallel Benchmarks in Java Michael A. Frumkin, Matthew Schultz, Haoqiang Jin, and Jerry 'fan NASA Advanced Supercomputing (NAS) Division NASA Ames Research Center, Moffett Field, CA 94035-1000 {frumkin, hj in, yan}_nas, nasa. gov Abstract Several features make Java an attractive choice for scientific applications. In order to gauge the applicability of Java to Computational Fluid Dynamics (CFD), we have implemented the NAS Parallel Benchmarks in Java. The performance and sealability of the benchmarks point out the areas where improvement in Java compiler tech- nology and in Java thread implementation would position Java closer to Fortran in the competition for scientific applications. 1 Introduction The portability, expressiveness, and safety of the Java language, supported by rapid progress in Java compiler technology, have created an interest in the High Performance Computing (HPC) community to evaluate Java on computationally intensive problems [8]. Java threads, RMI, and networking capabilities position Java well for programming on Shared Memory Parallel (SMP) computers and on computational grids. On the other hand issues of safety, lack of light weight objects, intermediate byte code interpretation, and array access overheads create challenges in achieving high performance for Java codes. The challenges are being addressed by work on implementation of efficient Java compilers [10] and by extending Java with classes implementing the data types used in HPC [9]. In this paper, we describe an implementation of the NAS Parallel Benchmarks (NPB) [1] in Java. The benchmark suite is accepted by the HPC community as an instrument for evaluating performance of parallel computers, compilers, and tools.
    [Show full text]
  • High Performance Computing: Models, Methods, & Means Benchmarking
    Prof. Thomas Sterling Department of Computer Science Louisiana State University January 27, 2011 HIGH PERFORMANCE COMPUTING: MODELS, METHODS, & MEANS BENCHMARKING CSC 7600 Lecture 4: Benchmarking, Spring 2011 Topics • Definitions, properties and applications • Early benchmarks • Linpack • Other parallel benchmarks • Organized benchmarking • Presentation and interpretation of results • Summary CSC 7600 Lecture 4: Benchmarking, 2 Spring 2011 Topics • Definitions, properties and applications • Early benchmarks • Linpack • Other parallel benchmarks • Organized benchmarking • Presentation and interpretation of results • Summary CSC 7600 Lecture 4: Benchmarking, 3 Spring 2011 Basic Performance Metrics • Time related: • Effectiveness: – Execution time [seconds] – Efficiency [%] • wall clock time • Sustained perf/peak perf • system and user time – Memory consumption – Latency [bytes] – Response time – Productivity [utility/($*second)] • Rate related: – Rate of computation • Performance measures: • floating point operations – Sustained Performance per second [flops] – Peak Performance • integer operations per – Benchmark sustained perf second [ops] • HPL Rmax – Data transfer (I/O) rate [bytes/second] CSC 7600 Lecture 4: Benchmarking, 4 Spring 2011 What Is a Benchmark? Benchmark: a standardized problem or test that serves as a basis for evaluation or comparison (as of computer system performance) [Merriam- Webster] • The term “benchmark” also commonly applies to specially-designed programs used in benchmarking • A benchmark should: – be domain
    [Show full text]
  • The NAS Parallel Benchmarks
    TITLE: The NAS Parallel Benchmarks AUTHOR: David H Bailey1 ACRONYMS: NAS, NPB DEFINITION: The NAS Parallel Benchmarks (NPB) are a suite of parallel computer per- formance benchmarks. They were originally developed at the NASA Ames Re- search Center in 1991 to assess high-end parallel supercomputers [?]. Although they are no longer used as widely as they once were for comparing high-end sys- tem performance, they continue to be studied and analyzed a great deal in the high-performance computing community. The acronym \NAS" originally stood for the Numerical Aeronautical Simulation Program at NASA Ames. The name of this organization was subsequently changed to the Numerical Aerospace Sim- ulation Program, and more recently to the NASA Advanced Supercomputing Center, although the acronym remains \NAS." The developers of the original NPB suite were David H. Bailey, Eric Barszcz, John Barton, David Browning, Russell Carter, LeoDagum, Rod Fatoohi, Samuel Fineberg, Paul Frederickson, Thomas Lasinski, Rob Schreiber, Horst Simon, V. Venkatakrishnan and Sisira Weeratunga. DISCUSSION: The original NAS Parallel Benchmarks consisted of eight individual bench- mark problems, each of which focused on some aspect of scientific computing. The principal focus was in computational aerophysics, although most of these benchmarks have much broader relevance, since in a much larger sense they are typical of many real-world scientific computing applications. The NPB suite grew out of the need for a more rational procedure to select new supercomputers for acquisition by NASA. The emergence of commercially available highly parallel computer systems in the late 1980s offered an attrac- tive alternative to parallel vector supercomputers that had been the mainstay of high-end scientific computing.
    [Show full text]
  • Data Extraction Tool
    TASK PARALLEL IMPLEMENTATION OF NAS PARALLEL BENCHMARKS Shashi Kumar Nanjaiah B.E, Visveswaraiah Technological University, Karnataka, India, 2005 PROJECT Submitted in partial satisfaction of the requirements for the degree of MASTER OF SCIENCE in COMPUTER SCIENCE at CALIFORNIA STATE UNIVERSITY, SACRAMENTO SPRING 2010 TASK PARALLEL IMPLEMENTATION OF NAS PARALLEL BENCHMARKS A Project by Shashi Kumar Nanjaiah Approved by: __________________________________, Committee Chair Dr. Chung E Wang __________________________________, Second Reader Dr. Ted Krovetz ____________________________ Date ii Student: Shashi Kumar Nanjaiah I certify that this student has met the requirements for format contained in the University format manual, and that this project is suitable for shelving in the Library and credit is to be awarded for the Project. __________________________, Graduate Coordinator ________________ Dr. Cui Zhang Date Department of Computer Science iii Abstract of TASK PARALLEL IMPLEMENTATION OF NAS PARALLEL BENCHMARKS by Shashi Kumar Nanjaiah The multi-core era brings new challenges to the programming community. Parallelization requirements of applications in mainstream computing and applications in emergent fields of high performance computing, such as informatics, must be explored. With parallelism now ubiquitous, programmability, composability, and reuse need to be closely examined in applications developed using existing parallel programming tools. Of the available programming models, task parallelism is the most promising and is able to parallelize regular and irregular applications alike [1, 2, 3]. In order to widely adopt task parallelism it is necessary for a tool to be configurable and extensible without any runtime penalties. Recently, PFunc [4], a novel library for expressing shared-memory task parallelism in C and C++ has been released to the open-source community.
    [Show full text]
  • High-Performance Conjugate-Gradient Benchmark
    Original Article The International Journal of High Performance Computing Applications 1–8 High-performance conjugate-gradient Ó The Author(s) 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav benchmark: A new metric for ranking DOI: 10.1177/1094342015593158 high-performance computing systems hpc.sagepub.com Jack Dongarra1, Michael A Heroux2 and Piotr Luszczek3 Abstract We describe a new high-performance conjugate-gradient (HPCG) benchmark. HPCG is composed of computations and data-access patterns commonly found in scientific applications. HPCG strives for a better correlation to existing codes from the computational science domain and to be representative of their performance. HPCG is meant to help drive the computer system design and implementation in directions that will better impact future performance improvement. Keywords Preconditioned conjugate gradient, multigrid smoothing, additive Schwarz, HPC benchmarking, validation and verification 1Introduction in its purpose to HPL (Dongarra et al., 2003) which is currently used to rank systems as part of the TOP500 Many aspects of the physical world may be modeled project (Meuer et al., 2013), but HPCG is intended to with partial differential equations (PDEs) and lend a better represent how today’s applications perform. hand to predictive capability so as to aid scientific dis- HPCGgeneratesaregularsparselinearsystemthatis covery and engineering optimization. The high- mathematically similar to a finite element, finite volume or performance conjugate-gradient (HPCG) benchmark is finite difference discretization of a three-dimensional heat used to test a high-performance conjugate (HPC) diffusion equation on a semi-regular grid. The problem is machine’s ability to solve these important scientific solved using domain decomposition (Smith et al., 1996) problems.
    [Show full text]