Dynamic buffer management to improve the performance of fault tolerance adaptive Network-On-Chip applications

S. Umamaheswari, N. Nancy, J. Perinbam
{"title":"Dynamic buffer management to improve the performance of fault tolerance adaptive Network-On-Chip applications","authors":"S. Umamaheswari, N. Nancy, J. Perinbam","doi":"10.1109/ICRTIT.2012.6206830","DOIUrl":null,"url":null,"abstract":"Networks-On-Chip are developed with a trade-off between latency and power dissipation defined at design time. But, if the communication pattern is changed, decisions taken at design time (say buffer size) may result in large area and power consumption or higher latency. Using large buffers to guarantee performance leads to excessive power dissipation. Small buffers reduce power consumption but result in increase in latency. The purpose of the proposed work is to design a heterogeneous router where the buffer slots are dynamically assigned to improve the performance, under different communication needs in fault-tolerant adaptive NoC applications. In the proposed router, buffer slots can dynamically be re-allocated for various applications to improve performance. Reallocation is based on the number of hotspots using EBLA (Extended Buffer Loan Algorithm). By introducing oversized IPs (OIPs), regular mesh-based NoC architecture may be destroyed. Resulting mesh-based NoC becomes irregular and needs new routing algorithms to solve routing problems in case of faulty links. A NoC with irregular 2D mesh topology is considered and an fault tolerant adaptive routing algorithm is used.","PeriodicalId":191151,"journal":{"name":"2012 International Conference on Recent Trends in Information Technology","volume":"84 2 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2012-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2012 International Conference on Recent Trends in Information Technology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICRTIT.2012.6206830","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3

Abstract

Networks-On-Chip are developed with a trade-off between latency and power dissipation defined at design time. But, if the communication pattern is changed, decisions taken at design time (say buffer size) may result in large area and power consumption or higher latency. Using large buffers to guarantee performance leads to excessive power dissipation. Small buffers reduce power consumption but result in increase in latency. The purpose of the proposed work is to design a heterogeneous router where the buffer slots are dynamically assigned to improve the performance, under different communication needs in fault-tolerant adaptive NoC applications. In the proposed router, buffer slots can dynamically be re-allocated for various applications to improve performance. Reallocation is based on the number of hotspots using EBLA (Extended Buffer Loan Algorithm). By introducing oversized IPs (OIPs), regular mesh-based NoC architecture may be destroyed. Resulting mesh-based NoC becomes irregular and needs new routing algorithms to solve routing problems in case of faulty links. A NoC with irregular 2D mesh topology is considered and an fault tolerant adaptive routing algorithm is used.
动态缓冲区管理,以提高自适应片上网络应用的容错性能
片上网络是在设计时定义的延迟和功耗之间进行权衡的。但是,如果通信模式发生了变化,那么在设计时所做的决定(比如缓冲区大小)可能会导致更大的面积和功耗,或者更高的延迟。使用大的缓冲区来保证性能会导致过大的功耗。较小的缓冲区可以降低功耗,但会导致延迟增加。在容错自适应NoC应用中,针对不同的通信需求,设计了一种动态分配缓冲槽的异构路由器,以提高性能。在所提出的路由器中,缓冲槽可以动态地重新分配给不同的应用程序,以提高性能。重新分配是基于热点的数量使用EBLA(扩展缓冲贷款算法)。引入超大ip (oip)可能会破坏常规的基于网格的NoC架构。导致基于网格的NoC变得不规则,需要新的路由算法来解决故障链路下的路由问题。考虑了具有不规则二维网格拓扑结构的NoC,采用了容错自适应路由算法。
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