05715nam 22007454a 450 991102013560332120200520144314.0978661031131697812803113141280311312978047036083504703608369780471732716047173271097804717327090471732702(CKB)1000000000355518(EBL)244302(OCoLC)166583856(SSID)ssj0000170844(PQKBManifestationID)11153514(PQKBTitleCode)TC0000170844(PQKBWorkID)10224531(PQKB)10774135(MiAaPQ)EBC244302(Perlego)2769251(EXLCZ)99100000000035551820041126d2006 uy 0engur|n|---|||||txtccrHigh performance computing paradigm and infrastructure /edited by Laurence T. Yang, Minyi GuoHoboken, N.J. J. Wileyc20061 online resource (818 p.)Wiley series on parallel and distributed computingDescription based upon print version of record.9780471654711 047165471X Includes bibliographical references and index.HIGH-PERFORMANCE COMPUTING; Contents; Preface; Contributors; PART 1 Programming Model; 1 ClusterGOP: A High-Level Programming Environment for Clusters; 1.1 Introduction; 1.2 GOP Model and ClusterGOP Architecture; 1.2.1 The ClusterGOP Architecture; 1.3 VisualGOP; 1.4 The ClusterGOP Library; 1.5 MPMD programming Support; 1.6 Programming Using ClusterGOP; 1.6.1 Support for Program Development; 1.6.2 Performance of ClusterGOP; 1.7 Summary; 2 The Challenge of Providing A High-Level Programming Model for High-Performance Computing; 2.1 Introduction; 2.2 HPC Architectures2.2.1 Early Parallel Processing Platforms2.2.2 Current HPC Systems; 2.3 HPC Programming Models: The First Generation; 2.3.1 The Message Passing Interface (MPI); 2.3.2 High Performance Fortran (HPF); 2.4 The Second Generation of HPC Programming Models; 2.4.1 OpenMP; 2.4.2 Other Shared-Memory APIs; 2.4.3 Is A Standard High-Level API for HPC in Sight?; 2.5 OpenMP for DMPs; 2.5.1 A Basic Translation to GA; 2.5.2 Implementing Sequential Regions; 2.5.3 Data and Work Distribution in GA; 2.5.4 Irregular Computation Example; 2.6 Experiments with OpenMP on DMPs; 2.7 Conclusions3 SAT: Toward Structured Parallelism Using Skeletons3.1 Introduction; 3.2 SAT: A Methodology Outline; 3.2.1 Motivation and Methodology; 3.2.2 Abstraction View: Basic Skeletons and Compositions; 3.2.3 Performance View: Collective Operations; 3.2.4 SAT: Combining Abstraction with Performance; 3.3 Skeletons and Collective Operations; 3.3.1 The H Skeleton and Its Standard Implementation; 3.3.2 Transformations for Performance View; 3.4 Case Study: Maximum Segment SUM (MSS); 3.5 Performance Aspect in SAT; 3.5.1 Performance Predictability; 3.5.2 Absolute Performance; 3.6 Conclusions and Related Work4 Bulk-Synchronous Parallelism: An Emerging Paradigm of High-Performance Computing4.1 The BSP Model; 4.1.2 BSP Versus Traditional Parallelism; 4.1.3 Memory Efficiency; 4.1.4 Memory Management; 4.1.5 Heterogeneity; 4.1.6 Subset Synchronization; 4.1.7 Other Variants of BSP; 4.2 BSP Programming; 4.2.1 The BSPlib Standard; 4.2.2 Beyond BSPlib; 4.3 Conclusions; 5 Cilk Versus MPI: Comparing Two Parallel Programming Styles on Heterogeneous Systems; 5.1 Introduction; 5.1.1 Message-Passing Run-Time Systems; 5.1.2 Cilk's Dataflow Model; 5.1.3 Terminology; 5.2 Experiments; 5.2.1 Programs; 5.2.2 Test Bed5.3 Results5.3.1 Fibonacci; 5.3.2 Traveling Salesman Problem; 5.3.3 N-Queens Problem; 5.3.4 Matrix Multiplication; 5.3.5 Finite Differencing; 5.3.6 Program Complexity; 5.4 Conclusion; 6 Nested Parallelism and Pipelining in OpenMP; 6.1 Introduction; 6.2 OpenMP Extensions for Nested Parallelism; 6.2.1 Parallelism Definition; 6.2.2 Thread Groups; 6.2.3 Evaluation of the Proposal; 6.3 OpenMP Extensions For Thread Synchronization; 6.3.1 Precedence Relations; 6.3.2 Evaluation of the Proposal; 6.4 Summary; 7 OpenMP for Chip Multiprocessors; 7.1 Introduction; 7.2 3SoC Architecture Overview7.2.1 QuadsThe state of the art of high-performance computingProminent researchers from around the world have gathered to present the state-of-the-art techniques and innovations in high-performance computing (HPC), including:* Programming models for parallel computing: graph-oriented programming (GOP), OpenMP, the stages and transformation (SAT) approach, the bulk-synchronous parallel (BSP) model, Message Passing Interface (MPI), and Cilk* Architectural and system support, featuring the code tiling compiler technique, the MigThread application-level migration and checkpointing package, thWiley series on parallel and distributed computing.High performance computingParallel processing (Electronic computers)Electronic data processingDistributed processingHigh performance computing.Parallel processing (Electronic computers)Electronic data processingDistributed processing.004/.35Yang Laurence Tianruo296155Guo Minyi968002MiAaPQMiAaPQMiAaPQBOOK9911020135603321High performance computing4417124UNINA