对容错并行处理器操作系统的概念和性能测量进行概述

C.A. Babikyan
{"title":"对容错并行处理器操作系统的概念和性能测量进行概述","authors":"C.A. Babikyan","doi":"10.1109/DASC.1990.111316","DOIUrl":null,"url":null,"abstract":"It is pointed out that mission critical applications of the future will require a computing system capable of high throughput as well as very high reliability. The fault tolerant parallel processor (FTPP), a system designed specifically to satisfy these goals, is described. The FTPP architecture consists of interconnection network/redundancy management hardware and standard commercial processors. The architecture provides flexibility in the appropriate balance of throughput and reliability for a given application. Furthermore, to maintain a system of high reliability the FTPP expeditiously identifies faulty components and performs some remedial operations. These redundancy management functions are performed by the operating system to relive the application from the knowledge of the underlying fault tolerance. How the operating system achieves redundancy management in conjunction with the fault tolerant hardware is described. Performance data to characterize system behavior are presented. Performance measurements indicate that the cost of fault tolerance does not significantly penalize forming redundancy management functions requires a mere .93 ms/frame more than a simplex processor performing no redundancy management.<<ETX>>","PeriodicalId":141205,"journal":{"name":"9th IEEE/AIAA/NASA Conference on Digital Avionics Systems","volume":"49 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1990-10-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"10","resultStr":"{\"title\":\"The fault tolerant parallel processor operating system concepts and performance measurement overview\",\"authors\":\"C.A. Babikyan\",\"doi\":\"10.1109/DASC.1990.111316\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"It is pointed out that mission critical applications of the future will require a computing system capable of high throughput as well as very high reliability. The fault tolerant parallel processor (FTPP), a system designed specifically to satisfy these goals, is described. The FTPP architecture consists of interconnection network/redundancy management hardware and standard commercial processors. The architecture provides flexibility in the appropriate balance of throughput and reliability for a given application. Furthermore, to maintain a system of high reliability the FTPP expeditiously identifies faulty components and performs some remedial operations. These redundancy management functions are performed by the operating system to relive the application from the knowledge of the underlying fault tolerance. How the operating system achieves redundancy management in conjunction with the fault tolerant hardware is described. Performance data to characterize system behavior are presented. Performance measurements indicate that the cost of fault tolerance does not significantly penalize forming redundancy management functions requires a mere .93 ms/frame more than a simplex processor performing no redundancy management.<<ETX>>\",\"PeriodicalId\":141205,\"journal\":{\"name\":\"9th IEEE/AIAA/NASA Conference on Digital Avionics Systems\",\"volume\":\"49 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"1990-10-15\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"10\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"9th IEEE/AIAA/NASA Conference on Digital Avionics Systems\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/DASC.1990.111316\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"9th IEEE/AIAA/NASA Conference on Digital Avionics Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/DASC.1990.111316","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 10

摘要

指出,未来的关键任务应用将需要具有高吞吐量和高可靠性的计算系统。本文描述了一种专门为满足这些目标而设计的容错并行处理器(FTPP)。FTPP架构由互连网络/冗余管理硬件和标准商用处理器组成。该体系结构为给定应用程序提供了适当平衡吞吐量和可靠性的灵活性。此外,为了保持系统的高可靠性,FTPP快速识别故障部件并执行一些补救操作。这些冗余管理功能由操作系统执行,以便从底层容错知识中重新体验应用程序。描述了操作系统如何结合容错硬件实现冗余管理。给出了表征系统行为的性能数据。性能测量表明,容错的成本不会显著影响形成冗余管理功能,只需要比不执行冗余管理的单形处理器多0.93 ms/帧。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
The fault tolerant parallel processor operating system concepts and performance measurement overview
It is pointed out that mission critical applications of the future will require a computing system capable of high throughput as well as very high reliability. The fault tolerant parallel processor (FTPP), a system designed specifically to satisfy these goals, is described. The FTPP architecture consists of interconnection network/redundancy management hardware and standard commercial processors. The architecture provides flexibility in the appropriate balance of throughput and reliability for a given application. Furthermore, to maintain a system of high reliability the FTPP expeditiously identifies faulty components and performs some remedial operations. These redundancy management functions are performed by the operating system to relive the application from the knowledge of the underlying fault tolerance. How the operating system achieves redundancy management in conjunction with the fault tolerant hardware is described. Performance data to characterize system behavior are presented. Performance measurements indicate that the cost of fault tolerance does not significantly penalize forming redundancy management functions requires a mere .93 ms/frame more than a simplex processor performing no redundancy management.<>
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:604180095
Book学术官方微信