Strongly consistent global state detection for on-line control of distributed applications

J. Borkowski
{"title":"Strongly consistent global state detection for on-line control of distributed applications","authors":"J. Borkowski","doi":"10.1109/EMPDP.2004.1271437","DOIUrl":null,"url":null,"abstract":"Global states can be used for distributed/parallel application monitoring and control. Strongly consistent global states (SCGS) are especially well suited for on-line controlling. Existing SCGS detecting algorithms work with process local states. They must wait for state terminations before complete states can be taken into account. Because of that the global states seen by a monitor always belong to the past. We present an algorithm, which works with unterminated local states. This approach lets the monitor detect SCGS sooner: currently lasting global states can be perceived promptly after they started. Application control based on SCGS detection should react quicker to arising situations when using the new algorithm. The quick reactions contribute to a better parallel/distributed application performance. Simulation tests confirm these suppositions. Our solution utilizes bounded maximal message transfer time. It is compared with another method, which employs frequent confirmation messages. While both methods can lead to similar application performance, our approach induces a few times lower both monitor load and network traffic.","PeriodicalId":105726,"journal":{"name":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2004-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"11","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"12th Euromicro Conference on Parallel, Distributed and Network-Based Processing, 2004. Proceedings.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/EMPDP.2004.1271437","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 11

Abstract

Global states can be used for distributed/parallel application monitoring and control. Strongly consistent global states (SCGS) are especially well suited for on-line controlling. Existing SCGS detecting algorithms work with process local states. They must wait for state terminations before complete states can be taken into account. Because of that the global states seen by a monitor always belong to the past. We present an algorithm, which works with unterminated local states. This approach lets the monitor detect SCGS sooner: currently lasting global states can be perceived promptly after they started. Application control based on SCGS detection should react quicker to arising situations when using the new algorithm. The quick reactions contribute to a better parallel/distributed application performance. Simulation tests confirm these suppositions. Our solution utilizes bounded maximal message transfer time. It is compared with another method, which employs frequent confirmation messages. While both methods can lead to similar application performance, our approach induces a few times lower both monitor load and network traffic.
用于分布式应用程序在线控制的强一致全局状态检测
全局状态可用于分布式/并行应用程序的监视和控制。强一致全局状态(SCGS)特别适合于在线控制。现有的SCGS检测算法适用于进程局部状态。它们必须等待状态终止,然后才能考虑完整状态。正因为如此,监视器看到的全局状态总是属于过去的。我们提出了一种算法,它适用于不终止的局部状态。这种方法可以让监控器更快地检测到SCGS:当前持续的全局状态可以在它们启动后立即被感知到。在使用新算法时,基于SCGS检测的应用程序控制应该对出现的情况做出更快的反应。快速反应有助于提高并行/分布式应用程序的性能。模拟试验证实了这些假设。我们的解决方案利用有界的最大消息传输时间。并与另一种使用频繁确认信息的方法进行了比较。虽然这两种方法都可以导致类似的应用程序性能,但我们的方法可以将监视器负载和网络流量降低几倍。
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
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:604180095
Book学术官方微信