Eliminating redundant communication of code generation for distributed memory machines

被引：0
|
作者：
Shen, Ya Nan [1 ]
Zhao, Rong Cai [1 ]
Wang, Lei [2 ]
机构：
[1] Natl Digital Switching Syst Engn & Technol R&D Ct, Zhengzhou 450002, Peoples R China
[2] Beijing Univ Aeronaut & Astronaut, Beijing 100083, Peoples R China
来源：
SNPD 2007: EIGHTH ACIS INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING, ARTIFICIAL INTELLIGENCE, NETWORKING, AND PARALLEL/DISTRIBUTED COMPUTING, VOL 2, PROCEEDINGS | 2007年
关键词：
D O I：
10.1109/SNPD.2007.482
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
How to, eliminate redundant communications is a key issue in developing, parallelizing compilers for distributed' memory, machines. An important way, to) improve the, performance of the code generated by parallelizing compiler in distributed memory systems is to eliminate as much communications as possible. This paper investigates algorithms for generating communications code from previously published work and modifies them to eliminate redundant communications. Based on the mathematical model of symbolic coefficient linear inequalities the revisited formal specification of communications sets are presented and the corresponding improved algorithms are given. Experimental results indicate that the new algorithms can reduce communication messages and improve the performance of parallelized code significantly.
引用
收藏
页码：751 / +
页数：2
相关论文
共 50 条
  • [31] Redundant computation partition on distributed-memory systems
    Chen, L
    Zhang, ZQ
    Feng, XB
    [J]. FIFTH INTERNATIONAL CONFERENCE ON ALGORITHMS AND ARCHITECTURES FOR PARALLEL PROCESSING, PROCEEDINGS, 2002, : 252 - 260
  • [32] Composable code generation for distributed Giotto
    Henzinger, TA
    Kirsch, CA
    Matic, S
    [J]. ACM SIGPLAN NOTICES, 2005, 40 (07) : 21 - 30
  • [33] Efficient index set generation for compiling HPF array statements on distributed-memory machines
    Kaushik, SD
    Huang, CH
    Sadayappan, P
    [J]. JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 1996, 38 (02) : 237 - 247
  • [34] ELIMINATING MEMORY-FRAGMENTATION WITHIN PARTITIONABLE SIMD SPMD MACHINES
    NICHOLS, MA
    SIEGEL, HJ
    DIETZ, HG
    QUONG, RW
    NATION, WG
    [J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 1991, 2 (03) : 290 - 303
  • [35] An overview of the APC compiler for distributed memory machines
    Zotov, S
    [J]. PDPTA'03: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOLS 1-4, 2003, : 1941 - 1945
  • [36] Load balancing strategies for distributed memory machines
    Diekmann, R
    Monien, B
    Preis, R
    [J]. MULTISCALE PHENOMENA AND THEIR SIMULATION, 1997, : 255 - 266
  • [37] A programming environment for heterogenous distributed memory machines
    Arapov, D
    Kalinov, A
    Lastovetsky, A
    Ledovskih, I
    Lewis, T
    [J]. SIXTH HETEROGENEOUS COMPUTING WORKSHOP (HCW '97), PROCEEDINGS, 1997, : 32 - 45
  • [38] On scheduling dynamic problems on distributed memory machines
    Sun, SX
    Zheng, WX
    [J]. PROCEEDINGS OF FOURTH INTERNATIONAL WORKSHOP ON CSCW IN DESIGN, 1999, : 387 - 389
  • [39] Cyclic reduction on distributed shared memory machines
    Allmann, S
    Rauber, T
    Rünger, G
    [J]. NINTH EUROMICRO WORKSHOP ON PARALLEL AND DISTRIBUTED PROCESSING, PROCEEDINGS, 2001, : 290 - 297
  • [40] Parallel numerical algorithms for distributed memory machines
    Bassomo, P
    Sakho, I
    Corbel, A
    [J]. PARALLEL COMPUTATION, 1999, 1557 : 581 - 583
12345