Graphite: A distributed parallel simulator for multicores

HPCA - 16 2010 The Sixteenth International Symposium on High-Performance Computer Architecture Pub Date : 2010-04-01 DOI:10.1109/HPCA.2010.5416635

Jason E. Miller, H. Kasture, George Kurian, Charles Gruenwald, Nathan Beckmann, Christopher Celio, J. Eastep, A. Agarwal

{"title":"Graphite: A distributed parallel simulator for multicores","authors":"Jason E. Miller, H. Kasture, George Kurian, Charles Gruenwald, Nathan Beckmann, Christopher Celio, J. Eastep, A. Agarwal","doi":"10.1109/HPCA.2010.5416635","DOIUrl":null,"url":null,"abstract":"This paper introduces the Graphite open-source distributed parallel multicore simulator infrastructure. Graphite is designed from the ground up for exploration of future multi-core processors containing dozens, hundreds, or even thousands of cores. It provides high performance for fast design space exploration and software development. Several techniques are used to achieve this including: direct execution, seamless multicore and multi-machine distribution, and lax synchronization. Graphite is capable of accelerating simulations by distributing them across multiple commodity Linux machines. When using multiple machines, it provides the illusion of a single process with a single, shared address space, allowing it to run off-the-shelf pthread applications with no source code modification. Our results demonstrate that Graphite can simulate target architectures containing over 1000 cores on ten 8-core servers. Performance scales well as more machines are added with near linear speedup in many cases. Simulation slowdown is as low as 41× versus native execution.","PeriodicalId":368621,"journal":{"name":"HPCA - 16 2010 The Sixteenth International Symposium on High-Performance Computer Architecture","volume":"5 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"524","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"HPCA - 16 2010 The Sixteenth International Symposium on High-Performance Computer Architecture","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/HPCA.2010.5416635","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 524

Abstract

This paper introduces the Graphite open-source distributed parallel multicore simulator infrastructure. Graphite is designed from the ground up for exploration of future multi-core processors containing dozens, hundreds, or even thousands of cores. It provides high performance for fast design space exploration and software development. Several techniques are used to achieve this including: direct execution, seamless multicore and multi-machine distribution, and lax synchronization. Graphite is capable of accelerating simulations by distributing them across multiple commodity Linux machines. When using multiple machines, it provides the illusion of a single process with a single, shared address space, allowing it to run off-the-shelf pthread applications with no source code modification. Our results demonstrate that Graphite can simulate target architectures containing over 1000 cores on ten 8-core servers. Performance scales well as more machines are added with near linear speedup in many cases. Simulation slowdown is as low as 41× versus native execution.

查看原文本刊更多论文

Graphite:一个多核的分布式并行模拟器

本文介绍了石墨开源分布式并行多核模拟器基础架构。石墨是为探索未来包含数十、数百甚至数千核的多核处理器而从头开始设计的。它为快速设计空间探索和软件开发提供了高性能。为了实现这一点，使用了几种技术，包括:直接执行、无缝多核和多机器分发以及松散同步。Graphite能够通过将模拟分布在多个商用Linux机器上来加速模拟。当使用多台机器时，它提供了一个具有单个共享地址空间的单个进程的假象，允许它运行现成的pthread应用程序，而无需修改源代码。我们的结果表明，石墨可以在10个8核服务器上模拟包含超过1000个核的目标架构。在许多情况下，随着机器数量的增加，性能也会随着接近线性的加速而增长。与本机执行相比，模拟速度低至41倍。

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

HPCA - 16 2010 The Sixteenth International Symposium on High-Performance Computer Architecture

自引率

0.00%

发文量