自动化的价值:MPI应用迁移方法研究

Karolina Sarnowska-Upton, A. Grimshaw
{"title":"自动化的价值:MPI应用迁移方法研究","authors":"Karolina Sarnowska-Upton, A. Grimshaw","doi":"10.1109/TrustCom.2013.149","DOIUrl":null,"url":null,"abstract":"This paper presents a user study of the process of migrating MPI applications manually. The gathered data quantifies the scale of the challenge that researchers face when attempting to use shared computing resources. Migrating to one site took on average 2.5 hours where the majority of the time was spent on learning, compiling, and debugging. Less experienced researchers took almost 50% more time and overall spread their effort across more days. Consequently, they took on average three weeks to migrate to four sites. The study data is used to investigate the speedup that can be gained by automating the migration process. How much more efficient application migration can be when utilizing FEAM, the Framework for Efficient Application Migration, instead of just manual methods is modeled in terms of time. While an ideal solution for general automation of application migration would enable computations to run on any available compute resource with minimal user interaction and in a tuned manner, FEAM focuses on what can be automated without recompiling applications. The calculations predict that using FEAM results in a 1.6x speedup on average or just under an hour of effort saved per site. This equates to more than two day of saved effort per site. The results presented in this paper underline the value of automation in the context of application migration.","PeriodicalId":206739,"journal":{"name":"2013 12th IEEE International Conference on Trust, Security and Privacy in Computing and Communications","volume":"87 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"The Value of Automation: A Study of MPI Application Migration Methods\",\"authors\":\"Karolina Sarnowska-Upton, A. Grimshaw\",\"doi\":\"10.1109/TrustCom.2013.149\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper presents a user study of the process of migrating MPI applications manually. The gathered data quantifies the scale of the challenge that researchers face when attempting to use shared computing resources. Migrating to one site took on average 2.5 hours where the majority of the time was spent on learning, compiling, and debugging. Less experienced researchers took almost 50% more time and overall spread their effort across more days. Consequently, they took on average three weeks to migrate to four sites. The study data is used to investigate the speedup that can be gained by automating the migration process. How much more efficient application migration can be when utilizing FEAM, the Framework for Efficient Application Migration, instead of just manual methods is modeled in terms of time. While an ideal solution for general automation of application migration would enable computations to run on any available compute resource with minimal user interaction and in a tuned manner, FEAM focuses on what can be automated without recompiling applications. The calculations predict that using FEAM results in a 1.6x speedup on average or just under an hour of effort saved per site. This equates to more than two day of saved effort per site. The results presented in this paper underline the value of automation in the context of application migration.\",\"PeriodicalId\":206739,\"journal\":{\"name\":\"2013 12th IEEE International Conference on Trust, Security and Privacy in Computing and Communications\",\"volume\":\"87 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2013-07-16\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2013 12th IEEE International Conference on Trust, Security and Privacy in Computing and Communications\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/TrustCom.2013.149\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 12th IEEE International Conference on Trust, Security and Privacy in Computing and Communications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TrustCom.2013.149","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

本文介绍了手动迁移MPI应用程序的用户研究过程。收集到的数据量化了研究人员在尝试使用共享计算资源时所面临的挑战的规模。迁移到一个站点平均花费2.5小时,其中大部分时间花在学习、编译和调试上。经验不足的研究人员多花了近50%的时间,而且总体上花了更多的时间。因此,他们平均花了三周的时间迁移到四个地点。研究数据用于研究通过自动化迁移过程可以获得的加速。当利用FEAM(高效应用程序迁移框架)而不是手工方法时,应用程序迁移的效率可以提高多少,这是根据时间来建模的。虽然应用程序迁移的一般自动化的理想解决方案将使计算能够以最小的用户交互和调优的方式在任何可用的计算资源上运行,但FEAM关注的是在不重新编译应用程序的情况下可以自动化的内容。计算结果表明,使用FEAM平均可以为每个站点节省1.6倍的速度,或者不到1小时的工作量。这相当于为每个站点节省了超过两天的工作。本文给出的结果强调了自动化在应用程序迁移上下文中的价值。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
The Value of Automation: A Study of MPI Application Migration Methods
This paper presents a user study of the process of migrating MPI applications manually. The gathered data quantifies the scale of the challenge that researchers face when attempting to use shared computing resources. Migrating to one site took on average 2.5 hours where the majority of the time was spent on learning, compiling, and debugging. Less experienced researchers took almost 50% more time and overall spread their effort across more days. Consequently, they took on average three weeks to migrate to four sites. The study data is used to investigate the speedup that can be gained by automating the migration process. How much more efficient application migration can be when utilizing FEAM, the Framework for Efficient Application Migration, instead of just manual methods is modeled in terms of time. While an ideal solution for general automation of application migration would enable computations to run on any available compute resource with minimal user interaction and in a tuned manner, FEAM focuses on what can be automated without recompiling applications. The calculations predict that using FEAM results in a 1.6x speedup on average or just under an hour of effort saved per site. This equates to more than two day of saved effort per site. The results presented in this paper underline the value of automation in the context of application migration.
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
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