Transforming the adaptive irregular out-of-core applications for hiding communication and disk I/O

被引:0
|
作者
Hu, Changjun [1 ]
Yao, Guangli [1 ]
Wang, Jue [1 ]
Li, Jianjiang [1 ]
机构
[1] Univ Sci & Technol Beijing, Sch Informat Engn, Beijing 100083, Peoples R China
关键词
program transformation; iteration reordering; computation-communication overlap; computation-disk I/O overlap;
D O I
暂无
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In adaptive irregular out-of-core applications, communications and mass disk I/O operations occupy a large portion of the overall execution. This paper presents a program transformation scheme to enable overlap of communication, computation and disk I/O in this kind of applications. We take programs in inspector-executor model as starting point, and transform them to a pipeline fashion. By decomposing the inspector phase and reordering iterations, more overlap opportunities are efficiently utilized. In the experiments, our techniques are applied to two important applications i.e. Partial differential equation solver and Molecular dynamics problems. For these applications, versions employing our techniques are almost 30% faster than inspector-executor versions.
引用
收藏
页码:1200 / 1213
页数:14
相关论文
共 50 条
  • [1] Compiler-based I/O prefetching for out-of-core applications
    Brown, AD
    Mowry, TC
    Krieger, O
    [J]. ACM TRANSACTIONS ON COMPUTER SYSTEMS, 2001, 19 (02): : 111 - 170
  • [2] CLIP: A Disk I/O Focused Parallel Out-of-Core Graph Processing System
    Ai, Zhiyuan
    Zhang, Mingxing
    Wu, Yongwei
    Qian, Xuehai
    Chen, Kang
    Zheng, Weimin
    [J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2019, 30 (01) : 45 - 62
  • [3] Parallelization of irregular out-of-core applications for distributed-memory systems
    Brezany, P
    Choudhary, A
    Dang, M
    [J]. HIGH-PERFORMANCE COMPUTING AND NETWORKING, 1997, 1225 : 811 - 820
  • [4] Automatic compiler-inserted I/O prefetching for out-of-core applications
    Mowry, TC
    Demke, AK
    Krieger, O
    [J]. PROCEEDINGS OF THE SECOND SYMPOSIUM ON OPERATING SYSTEMS DESIGN AND IMPLEMENTATION (OSDI '96), 1996, : 3 - 17
  • [5] Disk resident arrays: An array-oriented I/O library for out-of-core computations
    Nieplocha, J
    Foster, I
    [J]. FRONTIERS '96 - THE SIXTH SYMPOSIUM ON FRONTIERS OF MASSIVELY PARALLEL COMPUTING, PROCEEDINGS, 1996, : 196 - 204
  • [6] Irregular and out-of-core parallel computing on clusters
    Brezany, P
    Bubak, M
    Malawski, M
    Zajac, K
    [J]. PARALLEL PROCESSING APPLIED MATHEMATICS, 2002, 2328 : 299 - 306
  • [7] Global I/O optimizations for out-of-core computations
    Kandemir, M
    Kandaswamy, M
    Choudhary, A
    [J]. FOURTH INTERNATIONAL CONFERENCE ON HIGH-PERFORMANCE COMPUTING, PROCEEDINGS, 1997, : 401 - 406
  • [8] The practice of I/O optimizations for out-of-core computation
    Tang, JQ
    Fang, BX
    Hu, MZ
    Zhang, HL
    [J]. PARALLEL AND DISTRIBUTED COMPUTING: APPLICATIONS AND TECHNOLOGIES, PROCEEDINGS, 2004, 3320 : 81 - 85
  • [9] I/O Chunking and Latency Hiding Approach for Out-of-core Sorting Acceleration using GPU and Flash NVM
    Sato, Hitoshi
    Mizote, Ryo
    Matsuoka, Satoshi
    Ogawa, Hirotaka
    [J]. 2016 IEEE INTERNATIONAL CONFERENCE ON BIG DATA (BIG DATA), 2016, : 398 - 403
  • [10] Squeezing out All the Value of Loaded Data: An Out-of-core Graph Processing System with Reduced Disk I/O
    Ai, Zhiyuan
    Zhang, Mingxing
    Wu, Yongwei
    Qian, Xuehai
    Chen, Kang
    Zheng, Weimin
    [J]. 2017 USENIX ANNUAL TECHNICAL CONFERENCE (USENIX ATC '17), 2017, : 125 - 137