Scalability analysis tools for SPMD message-passing parallel programs

Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems Pub Date : 1994-01-31 DOI:10.1109/MASCOT.1994.284425

S. Sarukkai

{"title":"Scalability analysis tools for SPMD message-passing parallel programs","authors":"S. Sarukkai","doi":"10.1109/MASCOT.1994.284425","DOIUrl":null,"url":null,"abstract":"Tools to study the scalability of parallel programs, as number of processors (p) executing the program and problem size (n) being solved are increased, are a critical component of performance debugging environments for parallel programs. Simulations and scalability metrics have been used to address this issue. Simulation can accurately predict the execution time of a program for a specific (n,p) pair. However, it suffers from the drawback that one needs to simulate the program for each (n,p) pair of interest. On the other hand, while scalability metrics express the program performance as functions of n and p, they have been targeted to specific applications and there are no tools to automatically obtain simple first order scalability trends for generic parallel programs. We address the issue of automatically obtaining scalability trends for a class of data-independent message passing SPMD parallel programs. We validate our approach by considering example parallel programs executed on the Intel iPSC/860 hypercube. We show that insight into the scalability of the program can be obtained, using this approach.<<ETX>>","PeriodicalId":288344,"journal":{"name":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","volume":"27 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1994-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"10","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MASCOT.1994.284425","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 10

Abstract

Tools to study the scalability of parallel programs, as number of processors (p) executing the program and problem size (n) being solved are increased, are a critical component of performance debugging environments for parallel programs. Simulations and scalability metrics have been used to address this issue. Simulation can accurately predict the execution time of a program for a specific (n,p) pair. However, it suffers from the drawback that one needs to simulate the program for each (n,p) pair of interest. On the other hand, while scalability metrics express the program performance as functions of n and p, they have been targeted to specific applications and there are no tools to automatically obtain simple first order scalability trends for generic parallel programs. We address the issue of automatically obtaining scalability trends for a class of data-independent message passing SPMD parallel programs. We validate our approach by considering example parallel programs executed on the Intel iPSC/860 hypercube. We show that insight into the scalability of the program can be obtained, using this approach.<>

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

用于SPMD消息传递并行程序的可伸缩性分析工具

随着执行程序的处理器数量(p)和要解决的问题大小(n)的增加，研究并行程序的可伸缩性的工具是并行程序性能调试环境的关键组成部分。模拟和可伸缩性指标已经被用来解决这个问题。仿真可以准确地预测特定(n,p)对程序的执行时间。然而，它的缺点是需要对每个(n,p)对感兴趣的对象模拟程序。另一方面，虽然可伸缩性指标将程序性能表示为n和p的函数，但它们是针对特定应用程序的，并且没有工具可以自动获得通用并行程序的简单一阶可伸缩性趋势。我们解决了一类独立于数据的消息传递SPMD并行程序自动获取可伸缩性趋势的问题。我们通过考虑在Intel iPSC/860超立方体上执行的示例并行程序来验证我们的方法。我们表明，使用这种方法可以获得对程序可扩展性的洞察力。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Proceedings of International Workshop on Modeling, Analysis and Simulation of Computer and Telecommunication Systems

自引率

0.00%

发文量

期刊最新文献

Time constrained message transmission in a LAN environment Approximate analysis of a multi-class open queueing network with class blocking and push-out Evaluating memory system performance of a large scale NUMA multiprocessor Efficient technique for performance analysis of locking protocols A maximum entropy analysis of the single server queue