Automated performance prediction for scalable parallel computing

被引：4
作者：
Clement, MJ [1 ]
Quinn, MJ [1 ]
机构：
[1] OREGON STATE UNIV,DEPT COMP SCI,CORVALLIS,OR 97331
来源：
PARALLEL COMPUTING | 1997年 / 23卷 / 10期
关键词：
performance prediction; parallel systems; message passing; scalability; performance modeling; experimental results;
D O I：
10.1016/S0167-8191(97)00066-5
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
Performance prediction is necessary in order to deal with multi-dimensional performance effects on parallel systems. The compiler-generated analytical model developed in this paper accounts for the effects of cache behavior, CPU execution time and message passing overhead for real programs written in high level data-parallel languages. The performance prediction technique is shown to be effective in analyzing several non-trivial data-parallel applications as the problem size and number of processors vary. We leverage technology from the Maple symbolic manipulation system and the S-PLUS statistical package in order to present users with critical performance information necessary for performance debugging, architectural enhancement and procurement of parallel systems. The usability of these results is improved through specifying confidence intervals as well as predicted execution times for data-parallel applications. (C) 1997 Elsevier Science B.V.
引用
收藏
页码：1405 / 1420
页数：16
相关论文
共 28 条
  • [1] [Anonymous], [No title captured]
  • [2] BALASUNDARAM V, 1991, SIGPLAN NOTICES, V26, P213, DOI 10.1145/109626.109647
  • [3] Chambers JM., 1992, Statistical models
  • [4] Clement M. J., 1993, Proceedings SUPERCOMPUTING '93, P886, DOI 10.1145/169627.169856
  • [5] CLEMENT MJ, 1995, P 28 HAW INT C SYST
  • [6] CLEMENT MJ, 1996, P SUPERCOMPUTING 96
  • [7] CLEMENT MJ, 1994, THESIS OREGON STATE
  • [8] *COMM PHYS MATH EN, 1993, GRAND CHALL 1993 HIG
  • [9] CROVELLA ME, 1994, P SUPERCOMPUTING 94
  • [10] FAHRINGER T, 1993, ACPCTR931 U VIENN DE
123