性能建模与预测框架

ACM/IEEE SC 2002 Conference (SC'02) Pub Date : 2002-11-16 DOI:10.1109/SC.2002.10004

A. Snavely, L. Carrington, N. Wolter, Jesús Labarta, Rosa M. Badia, A. Purkayastha

{"title":"性能建模与预测框架","authors":"A. Snavely, L. Carrington, N. Wolter, Jesús Labarta, Rosa M. Badia, A. Purkayastha","doi":"10.1109/SC.2002.10004","DOIUrl":null,"url":null,"abstract":"Cycle-accurate simulation is far too slow for modeling the expected performance of full parallel applications on large HPC systems. And just running an application on a system and observing wallclock time tells you nothing about why the application performs as it does (and is anyway impossible on yet-to-be-built systems). Here we present a framework for performance modeling and prediction that is faster than cycle-accurate simulation, more informative than simple benchmarking, and is shown useful for performance investigations in several dimensions.","PeriodicalId":302800,"journal":{"name":"ACM/IEEE SC 2002 Conference (SC'02)","volume":"38 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2002-11-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"263","resultStr":"{\"title\":\"A Framework for Performance Modeling and Prediction\",\"authors\":\"A. Snavely, L. Carrington, N. Wolter, Jesús Labarta, Rosa M. Badia, A. Purkayastha\",\"doi\":\"10.1109/SC.2002.10004\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Cycle-accurate simulation is far too slow for modeling the expected performance of full parallel applications on large HPC systems. And just running an application on a system and observing wallclock time tells you nothing about why the application performs as it does (and is anyway impossible on yet-to-be-built systems). Here we present a framework for performance modeling and prediction that is faster than cycle-accurate simulation, more informative than simple benchmarking, and is shown useful for performance investigations in several dimensions.\",\"PeriodicalId\":302800,\"journal\":{\"name\":\"ACM/IEEE SC 2002 Conference (SC'02)\",\"volume\":\"38 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2002-11-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"263\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"ACM/IEEE SC 2002 Conference (SC'02)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/SC.2002.10004\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACM/IEEE SC 2002 Conference (SC'02)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SC.2002.10004","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 263

摘要

周期精确的模拟对于模拟大型HPC系统上完全并行应用程序的预期性能来说太慢了。仅仅在系统上运行应用程序并观察时钟时间并不能告诉您为什么应用程序会这样执行(无论如何，在尚未构建的系统上是不可能的)。在这里，我们提出了一个性能建模和预测框架，它比周期精确的模拟更快，比简单的基准测试更有信息量，并且对几个维度的性能调查很有用。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

A Framework for Performance Modeling and Prediction

Cycle-accurate simulation is far too slow for modeling the expected performance of full parallel applications on large HPC systems. And just running an application on a system and observing wallclock time tells you nothing about why the application performs as it does (and is anyway impossible on yet-to-be-built systems). Here we present a framework for performance modeling and prediction that is faster than cycle-accurate simulation, more informative than simple benchmarking, and is shown useful for performance investigations in several dimensions.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

ACM/IEEE SC 2002 Conference (SC'02)

自引率

0.00%

发文量