Unified Programming Models for Heterogeneous High-Performance Computers

被引:0
|
作者
Zi-Xuan Ma
Yu-Yang Jin
Shi-Zhi Tang
Hao-Jie Wang
Wei-Cheng Xue
Ji-Dong Zhai
Wei-Min Zheng
机构
[1] Tsinghua University,Department of Computer Science and Technology
关键词
performance portability; programming model; heterogeneous supercomputer;
D O I
暂无
中图分类号
学科分类号
摘要
Unified programming models can effectively improve program portability on various heterogeneous high-performance computers. Existing unified programming models put a lot of effort to code portability but are still far from achieving good performance portability. In this paper, we present a preliminary design of a performance-portable unified programming model including four aspects: programming language, programming abstraction, compilation optimization, and scheduling system. Specifically, domain-specific languages introduce domain knowledge to decouple the optimizations for different applications and architectures. The unified programming abstraction unifies the common features of different architectures to support common optimizations. Multi-level compilation optimization enables comprehensive performance optimization based on multi-level intermediate representations. Resource-aware lightweight runtime scheduling system improves the resource utilization of heterogeneous computers. This is a perspective paper to show our viewpoints on programming models for emerging heterogeneous systems.
引用
收藏
页码:211 / 218
页数:7
相关论文
共 50 条
  • [41] Achieving advanced CFD capabilities with high-performance computers
    Masters, James
    Woeber, Carolyn
    AEROSPACE AMERICA, 2019, 57 (11) : 27 - 27
  • [42] High-performance computers: The next generation .D.
    Messina, P
    COMPUTERS IN PHYSICS, 1997, 11 (05): : 454 - 466
  • [43] HIGH-PERFORMANCE COMPUTING/COMPUTERS - SIMULATION MODELING AND APPLICATIONS
    OBAIDAT, MS
    SIMULATION, 1993, 61 (03) : 149 - 150
  • [44] Managing clusters of geographically distributed high-performance computers
    Brune, M
    Gehring, J
    Keller, A
    Reinefeld, A
    CONCURRENCY-PRACTICE AND EXPERIENCE, 1999, 11 (15): : 887 - 911
  • [45] INDUSTRIAL COMPUTERS PROVIDE CHOICE OF HIGH-PERFORMANCE PROCESSORS
    CLEAVELAND, P
    I&CS-INSTRUMENTATION & CONTROL SYSTEMS, 1994, 67 (10): : 134 - 134
  • [46] High-performance computers: The next generation (part II)
    Messina, P
    COMPUTERS IN PHYSICS, 1997, 11 (06): : 598 - 610
  • [47] High-performance special-purpose computers in science
    Fukushige, T
    Hut, P
    Makino, J
    COMPUTING IN SCIENCE & ENGINEERING, 1999, 1 (02) : 12 - +
  • [48] Reducing Computational Costs of Automatic Calibration of Rainfall-Runoff Models: Meta-Models or High-Performance Computers?
    Semiromi, Majid Taie
    Omidvar, Sorush
    Kamali, Bahareh
    WATER, 2018, 10 (10)
  • [49] High-performance programming frameworks for numerical simulation
    Zeyao Mo
    National Science Review, 2016, 3 (01) : 28 - 29
  • [50] Flux: A language for programming high-performance servers
    Burns, Brendan
    Grimaldi, Kevin
    Kostadinov, Alexander
    Berger, Emery D.
    Corner, Mark D.
    USENIX ASSOCIATION PROCEEDINGS OF THE 2006 USENIX ANNUAL TECHNICAL CONFERENCE, 2006, : 129 - +