High speed dynamic fault-tolerance

J. Sengupta, P. Bansal
{"title":"High speed dynamic fault-tolerance","authors":"J. Sengupta, P. Bansal","doi":"10.1109/TENCON.2001.949677","DOIUrl":null,"url":null,"abstract":"Multistage interconnection networks have been used as a favored ATM switch fabric. This paper analyzes the fault-tolerance and performance abilities of the proposed Phi network and compares it with its predecessor, the Omega network. The reduced numbers of stages of the designed MIN diminish the latency within the network markedly. The methods for routing permutations in the presence and absence of faulty components in both the networks have been analytically compared. The irregular nature of the Phi network allows 50% of the permutations to pass at the minimum path length of 2 whereas others pass at adaptable path length counting on the status of the path, largest being log/sub 2/N for a network of size N. The bandwidth of the new network shows optimal benefits. Bounds on reliability exhibit graceful degradation with time showing distinct gains over Omega where it reduces drastically, increasing the time between failures of the system. The comparison of these networks shows that irregular Phi network is a preferred choice to be used in the high-speed multiprocessor environment.","PeriodicalId":358168,"journal":{"name":"Proceedings of IEEE Region 10 International Conference on Electrical and Electronic Technology. TENCON 2001 (Cat. No.01CH37239)","volume":"59 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2001-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"19","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of IEEE Region 10 International Conference on Electrical and Electronic Technology. TENCON 2001 (Cat. No.01CH37239)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/TENCON.2001.949677","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 19

Abstract

Multistage interconnection networks have been used as a favored ATM switch fabric. This paper analyzes the fault-tolerance and performance abilities of the proposed Phi network and compares it with its predecessor, the Omega network. The reduced numbers of stages of the designed MIN diminish the latency within the network markedly. The methods for routing permutations in the presence and absence of faulty components in both the networks have been analytically compared. The irregular nature of the Phi network allows 50% of the permutations to pass at the minimum path length of 2 whereas others pass at adaptable path length counting on the status of the path, largest being log/sub 2/N for a network of size N. The bandwidth of the new network shows optimal benefits. Bounds on reliability exhibit graceful degradation with time showing distinct gains over Omega where it reduces drastically, increasing the time between failures of the system. The comparison of these networks shows that irregular Phi network is a preferred choice to be used in the high-speed multiprocessor environment.
高速动态容错
多级互连网络已被用作ATM交换结构的首选。本文分析了所提出的Phi网络的容错性和性能,并与它的前身Omega网络进行了比较。所设计的最小阶段数的减少显著降低了网络内的延迟。对两种网络中存在和不存在故障组件时的路由排列方法进行了分析比较。Phi网络的不规则特性允许50%的排列在最小路径长度为2时通过,而其他排列则根据路径的状态以自适应路径长度通过,对于大小为N的网络,最大的路径长度为log/sub 2/N。新网络的带宽显示出最佳效益。随着时间的推移,可靠性的界限表现出优雅的退化,在急剧减少的情况下表现出明显的增益,增加了系统故障之间的时间间隔。这些网络的比较表明,不规则Phi网络是高速多处理器环境下的首选网络。
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