The Omni OpenMP compiler on the distributed shared memory of Cenju-4

被引:0
|
作者
Kusano, K [1 ]
Sato, M
Hosomi, T
Seo, Y
机构
[1] NEC Corp Ltd, C&C Media Res Labs, Kawasaki, Kanagawa, Japan
[2] Real World Comp Partnership, Tsukuba, Ibaraki, Japan
关键词
D O I
暂无
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
This paper describes an implementation and a preliminary evaluation of the Omni OpenMP compiler on a parallel computer Cenju-4. The Cenju-4 is a parallel computer which support hardware distributed shared memory (DSM) system. The shared address space is explicitly allocated at the initialization phase of the program. The Omni converts all global variable declarations into indirect references through the pointers, and generates code to allocate those variables in the shared address space at runtime. The OpenMP programs can execute on a distributed memory machine with hardware DSM by using the Omni. The preliminary results using benchmark programs show that the performance of OpenMP programs didn't scales. While its performance of OpenMP benchmark programs scales poorly, it enables users to execute the same program on both a shared memory machine and a distribute memory machine.
引用
收藏
页码:20 / 30
页数:11
相关论文
共 50 条
  • [1] OpenMP compiler for distributed memory architectures
    WANG Jue
    [J]. Science China(Information Sciences), 2010, 53 (05) : 932 - 944
  • [2] OpenMP compiler for distributed memory architectures
    Wang Jue
    Hu ChangJun
    Zhang JiLin
    Li JianJiang
    [J]. SCIENCE CHINA-INFORMATION SCIENCES, 2010, 53 (05) : 932 - 944
  • [3] OpenMP compiler for distributed memory architectures
    Jue Wang
    ChangJun Hu
    JiLin Zhang
    JianJiang Li
    [J]. Science China Information Sciences, 2010, 53 : 932 - 944
  • [4] Architecture of a parallel computer Cenju-4
    Kanoh, Y
    Hirose, T
    Nakamura, M
    Hosomi, T
    Tatsukawa, K
    Araki, H
    Sugawara, T
    Nakata, T
    [J]. INNOVATIVE ARCHITECTURE FOR FUTURE GENERATION HIGH-PERFORMANCE PROCESSORS AND SYSTEMS, 1999, : 105 - 113
  • [5] Performance evaluation of the Omni OpenMP compiler
    Kusano, K
    Satoh, S
    Sato, M
    [J]. HIGH PERFORMANCE COMPUTING, PROCEEDINGS, 2000, 1940 : 403 - 414
  • [6] Architecture and the software environment of parallel computer Cenju-4
    Nakata, T
    Kanoh, Y
    Tatsukawa, K
    Yanagida, S
    Nishi, N
    Takayama, H
    [J]. NEC RESEARCH & DEVELOPMENT, 1998, 39 (04): : 385 - 390
  • [7] OMPCUDA : OpenMP Execution Framework for CUDA Based on Omni OpenMP Compiler
    Ohshima, Satoshi
    Hirasawa, Shoichi
    Honda, Hiroki
    [J]. BEYOND LOOP LEVEL PARALLELISM IN OPENMP: ACCELERATORS, TASKING AND MORE, PROCEEDINGS, 2010, 6132 : 161 - +
  • [8] An OpenMP Compiler for Efficient Use of Distributed Scratchpad Memory in MPSoCs
    Marongiu, Andrea
    Benini, Luca
    [J]. IEEE TRANSACTIONS ON COMPUTERS, 2012, 61 (02) : 222 - 236
  • [9] Optimizing OpenMP Programs on Software Distributed Shared Memory Systems
    Seung-Jai Min
    Ayon Basumallik
    Rudolf Eigenmann
    [J]. International Journal of Parallel Programming, 2003, 31 : 225 - 249
  • [10] OpenMP-oriented applications for distributed shared memory architectures
    Marowka, A
    Liu, ZY
    Chapman, B
    [J]. CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2004, 16 (04): : 371 - 384