SimplePipe: a simulation tool for task allocation and design of processor pipelines with application to network processors

M. Franklin, Vinayak Joshi
{"title":"SimplePipe: a simulation tool for task allocation and design of processor pipelines with application to network processors","authors":"M. Franklin, Vinayak Joshi","doi":"10.1109/MASCOT.2004.1348182","DOIUrl":null,"url":null,"abstract":"SimplePipe is a simulation framework/tool based on SimpleScalar (Austin, T. et al., IEEE Computer, 2002; http://www.simplescalar.com). SimplePipe analyzes the performance effects of alternative task allocations in systems of multiple pipelines where pipeline stages are either processors or dedicated hardware functions. Tasks are defined in terms of sequences of separate C program executions with each sequence representing the functional requirements of a flow. Performance effects associated with alternative assignment of shared memory modules to the pipeline stage can also be explored. SimplePipe was motivated by the performance requirements of network processors (NPs) that are configurable into multiple processor pipelines. The assignment of communication flow tasks to pipeline stages, selection of the number of stages, determination of processor cache sizes and the assignment of shared memory modules are important design decisions impacting performance. An assignment study illustrating SimplePipe capabilities is presented.","PeriodicalId":32394,"journal":{"name":"Performance","volume":"21 1","pages":"59-66"},"PeriodicalIF":0.0000,"publicationDate":"2004-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"5","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Performance","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/MASCOT.2004.1348182","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 5

Abstract

SimplePipe is a simulation framework/tool based on SimpleScalar (Austin, T. et al., IEEE Computer, 2002; http://www.simplescalar.com). SimplePipe analyzes the performance effects of alternative task allocations in systems of multiple pipelines where pipeline stages are either processors or dedicated hardware functions. Tasks are defined in terms of sequences of separate C program executions with each sequence representing the functional requirements of a flow. Performance effects associated with alternative assignment of shared memory modules to the pipeline stage can also be explored. SimplePipe was motivated by the performance requirements of network processors (NPs) that are configurable into multiple processor pipelines. The assignment of communication flow tasks to pipeline stages, selection of the number of stages, determination of processor cache sizes and the assignment of shared memory modules are important design decisions impacting performance. An assignment study illustrating SimplePipe capabilities is presented.
SimplePipe:一个用于任务分配和处理器管道设计的仿真工具,适用于网络处理器
SimplePipe是一个基于SimpleScalar的仿真框架/工具(Austin, T. et al., IEEE Computer, 2002;http://www.simplescalar.com)。SimplePipe分析了多个管道系统中可选任务分配的性能影响,其中管道阶段要么是处理器,要么是专用硬件功能。任务是根据单独的C程序执行序列来定义的,每个序列代表一个流的功能需求。还可以探索与将共享内存模块分配到管道阶段相关的性能影响。SimplePipe是由可配置到多个处理器管道的网络处理器(NPs)的性能需求驱动的。通信流任务分配到流水线阶段、阶段数量的选择、处理器缓存大小的确定以及共享内存模块的分配是影响性能的重要设计决策。介绍了一个作业研究,说明了SimplePipe的功能。
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 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信