Regional Congestion Aware Odd Even Routing with Fair Arbitration for Network on Chip

Rohith Somisetty, V. Karthik, M. R. S. Srujan, M. Vinodhini
{"title":"Regional Congestion Aware Odd Even Routing with Fair Arbitration for Network on Chip","authors":"Rohith Somisetty, V. Karthik, M. R. S. Srujan, M. Vinodhini","doi":"10.1109/icecct52121.2021.9616741","DOIUrl":null,"url":null,"abstract":"Advancement in current VLSI technologies increases the number of cores or Intellectual Properties (IPs) integrated in System on Chip (SoC) and this results in many-core and multi-core SoCs. Traditional interconnects used in these SoCs introduce multiple processing and traverse delays. In SoC, Network on Chip (NoC) overcomes the problem faced with traditional interconnect and provides an effective and efficient data transfer between the cores. Further in NoC, there is a growing need for routing algorithms which are flexible enough to provide minimal paths and congestion awareness. In this paper, a Regional Congestion Aware Odd Even routing algorithm with Fair Arbitration (RCAOE-FA) based on adaptive Odd Even (OE) turn model is proposed which provides deadlock freedom, on a 2D mesh topology based NoC. The proposed algorithm is implemented and analyzed using the cycle accurate simulator for four different traffic patterns namely – Random, Transpose-1, Transpose-2 and Bit-Reversal. The results of the same are compared with the predecessor algorithm under all the aforementioned traffic patterns. After implementing the algorithm it is observed that, the proposed routing algorithm shows considerable reduction in packet latency for transpose 1, transpose 2, bit reversal and random synthetic traffic patterns.","PeriodicalId":155129,"journal":{"name":"2021 Fourth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 Fourth International Conference on Electrical, Computer and Communication Technologies (ICECCT)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/icecct52121.2021.9616741","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 2

Abstract

Advancement in current VLSI technologies increases the number of cores or Intellectual Properties (IPs) integrated in System on Chip (SoC) and this results in many-core and multi-core SoCs. Traditional interconnects used in these SoCs introduce multiple processing and traverse delays. In SoC, Network on Chip (NoC) overcomes the problem faced with traditional interconnect and provides an effective and efficient data transfer between the cores. Further in NoC, there is a growing need for routing algorithms which are flexible enough to provide minimal paths and congestion awareness. In this paper, a Regional Congestion Aware Odd Even routing algorithm with Fair Arbitration (RCAOE-FA) based on adaptive Odd Even (OE) turn model is proposed which provides deadlock freedom, on a 2D mesh topology based NoC. The proposed algorithm is implemented and analyzed using the cycle accurate simulator for four different traffic patterns namely – Random, Transpose-1, Transpose-2 and Bit-Reversal. The results of the same are compared with the predecessor algorithm under all the aforementioned traffic patterns. After implementing the algorithm it is observed that, the proposed routing algorithm shows considerable reduction in packet latency for transpose 1, transpose 2, bit reversal and random synthetic traffic patterns.
基于公平仲裁的片上网络区域拥塞感知奇偶路由
当前VLSI技术的进步增加了集成在片上系统(SoC)中的核心或知识产权(ip)的数量,这导致了多核和多核SoC。在这些soc中使用的传统互连引入了多个处理和遍历延迟。在SoC中,片上网络(NoC)克服了传统互连所面临的问题,并在核心之间提供了有效和高效的数据传输。在NoC中,越来越需要足够灵活的路由算法,以提供最小的路径和拥塞感知。本文提出了一种基于自适应奇偶(OE)回合模型的具有公平仲裁的区域感知拥塞奇偶路由算法(RCAOE-FA),该算法在基于NoC的二维网格拓扑结构上提供了死锁自由。针对随机、转置-1、转置-2和位反转四种不同的流量模式,利用周期精确模拟器对该算法进行了实现和分析。在上述所有流量模式下,将其结果与前代算法进行了比较。在实现算法后,可以观察到所提出的路由算法在转置1、转置2、位反转和随机合成流量模式下的数据包延迟显著降低。
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