Efficient Network on Chip (NoC) using heterogeneous circuit switched routers

Anuja Naik, T. K. Ramesh
{"title":"Efficient Network on Chip (NoC) using heterogeneous circuit switched routers","authors":"Anuja Naik, T. K. Ramesh","doi":"10.1109/VLSI-SATA.2016.7593043","DOIUrl":null,"url":null,"abstract":"Network-on-Chip (NoC) architecture in recent years has been considered as the overwhelming communication solution to provide scalability in multi core systems over traditional bus-based communication architecture. There is an increased use of multi-core with NoC in embedded systems solutions. Energy efficiency in the Network-on-Chip (NoC) is one of the key challenges as these embedded systems are typically battery-powered. Router architecture impacts the performance of NoC. With increased interest towards circuit-switched routers, in this paper, we have proposed an area and energy efficient 5-port, 4 Lane circuit switched router using a CLOS network which presents the advantages of area and energy efficiency. To further improve energy efficiency of NoC, we use a hybrid architecture by mixing buffered and bufferless routers. Our results shows that by using CLOS switch network, we can gain 32% reduction in area and 26% reduction in power compared to Crossbar switch of the same size. Our comparison of using an 8×8 mesh heterogeneous router topology shows a reduction of 3% to 18% in silicon area and 10% to 15% in total power using bufferless routers compared to a fully buffered configuration.","PeriodicalId":328401,"journal":{"name":"2016 International Conference on VLSI Systems, Architectures, Technology and Applications (VLSI-SATA)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"12","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2016 International Conference on VLSI Systems, Architectures, Technology and Applications (VLSI-SATA)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/VLSI-SATA.2016.7593043","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 12

Abstract

Network-on-Chip (NoC) architecture in recent years has been considered as the overwhelming communication solution to provide scalability in multi core systems over traditional bus-based communication architecture. There is an increased use of multi-core with NoC in embedded systems solutions. Energy efficiency in the Network-on-Chip (NoC) is one of the key challenges as these embedded systems are typically battery-powered. Router architecture impacts the performance of NoC. With increased interest towards circuit-switched routers, in this paper, we have proposed an area and energy efficient 5-port, 4 Lane circuit switched router using a CLOS network which presents the advantages of area and energy efficiency. To further improve energy efficiency of NoC, we use a hybrid architecture by mixing buffered and bufferless routers. Our results shows that by using CLOS switch network, we can gain 32% reduction in area and 26% reduction in power compared to Crossbar switch of the same size. Our comparison of using an 8×8 mesh heterogeneous router topology shows a reduction of 3% to 18% in silicon area and 10% to 15% in total power using bufferless routers compared to a fully buffered configuration.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
采用异构电路交换路由器的高效片上网络(NoC)
近年来,片上网络(NoC)架构被认为是一种压倒传统的基于总线的通信架构的通信解决方案,以提供多核系统的可扩展性。嵌入式系统解决方案中越来越多地使用带有NoC的多核。由于这些嵌入式系统通常由电池供电,因此片上网络(NoC)的能源效率是关键挑战之一。路由器架构影响NoC的性能。随着对电路交换路由器的兴趣增加,在本文中,我们提出了一种区域和节能的5端口,4通道电路交换路由器,使用CLOS网络,具有区域和能源效率的优势。为了进一步提高NoC的能源效率,我们使用了一种混合架构,混合了缓冲和无缓冲路由器。结果表明,与同等尺寸的Crossbar开关相比,采用CLOS开关网络的开关面积可减少32%,功耗可减少26%。我们使用8×8网状异构路由器拓扑的比较表明,与完全缓冲配置相比,使用无缓冲路由器的硅面积减少了3%至18%,总功耗减少了10%至15%。
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 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Effect on temperature and time in parallel test scheduling with alterations in layers arrangements of 3D stacked SoCs A hardware optimized low power RNM compensated three stage operational amplifier with embedded capacitance multiplier compensation Reconfigurable side channel attack resistant true random number generator FPGA implementation of face recognition system using efficient 5/3 2D-lifting scheme Design of CMOS programmable output binary and fibonacci switched capacitor step-down DC-DC converter
×
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
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 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 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1