片上网络中的4-PAM互连,用于高吞吐量和对延迟敏感的应用

Ahmad Mansour, Ahmed El-Naggar, Bassma Al-Abassy, Mostafa Khamis, A. Shalaby
{"title":"片上网络中的4-PAM互连,用于高吞吐量和对延迟敏感的应用","authors":"Ahmad Mansour, Ahmed El-Naggar, Bassma Al-Abassy, Mostafa Khamis, A. Shalaby","doi":"10.1109/ISQED.2018.8357274","DOIUrl":null,"url":null,"abstract":"In this paper, a network-on-chip four-level pulse amplitude modulation (4-PAM) scheme is proposed to be used for communication within the network itself in MPSoCs. A current-mode based 4-PAM transmitter is used to encode data transactions between neighboring routers. Decoding data streams is done by a flash-ADC based receiver using clocked latched type comparators. Additionally, this scheme is implemented on networks utilizing high-radix routers with a local concentration factor of 2 IPs per node to encode data streams injected into the network at the network interface and decode them at the input port of the router. We also discuss the required modifications to the router architecture in the input port buffers and introduce a two-stage allocation method to resolve conflicts of output port requests which is essential to maintain system stability after saturation by utilizing a fair flow control methodology. This results in a reduction in wiring load for each router which is an added value that facilitates the routing stage. The evaluation is extended to reflect the overall network performance supporting the use of multi-valued logic and estimate the overhead of implementation on area and power budgets.","PeriodicalId":213351,"journal":{"name":"2018 19th International Symposium on Quality Electronic Design (ISQED)","volume":"15 6","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2018-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":"{\"title\":\"A 4-PAM interconnect in network-on-chip for high-throughput and latency-sensitive applications\",\"authors\":\"Ahmad Mansour, Ahmed El-Naggar, Bassma Al-Abassy, Mostafa Khamis, A. Shalaby\",\"doi\":\"10.1109/ISQED.2018.8357274\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this paper, a network-on-chip four-level pulse amplitude modulation (4-PAM) scheme is proposed to be used for communication within the network itself in MPSoCs. A current-mode based 4-PAM transmitter is used to encode data transactions between neighboring routers. Decoding data streams is done by a flash-ADC based receiver using clocked latched type comparators. Additionally, this scheme is implemented on networks utilizing high-radix routers with a local concentration factor of 2 IPs per node to encode data streams injected into the network at the network interface and decode them at the input port of the router. We also discuss the required modifications to the router architecture in the input port buffers and introduce a two-stage allocation method to resolve conflicts of output port requests which is essential to maintain system stability after saturation by utilizing a fair flow control methodology. This results in a reduction in wiring load for each router which is an added value that facilitates the routing stage. The evaluation is extended to reflect the overall network performance supporting the use of multi-valued logic and estimate the overhead of implementation on area and power budgets.\",\"PeriodicalId\":213351,\"journal\":{\"name\":\"2018 19th International Symposium on Quality Electronic Design (ISQED)\",\"volume\":\"15 6\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2018-03-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"1\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2018 19th International Symposium on Quality Electronic Design (ISQED)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISQED.2018.8357274\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2018 19th International Symposium on Quality Electronic Design (ISQED)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISQED.2018.8357274","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

摘要

本文提出了一种用于mpsoc网络内部通信的片上网络四电平脉冲幅度调制(4-PAM)方案。基于电流模式的4-PAM发送器用于对相邻路由器之间的数据事务进行编码。解码数据流由基于闪存adc的接收器使用时钟锁存型比较器完成。此外,该方案还在使用高基数路由器的网络上实现,该路由器的本地集中系数为每节点2个ip,通过网络接口对注入网络的数据流进行编码,并在路由器的输入端口进行解码。我们还讨论了输入端口缓冲区中路由器架构的必要修改,并介绍了一种两阶段分配方法来解决输出端口请求的冲突,这对于利用公平的流量控制方法在饱和后保持系统稳定性至关重要。这将减少每个路由器的布线负载,这是一个附加价值,有利于路由阶段。该评估扩展到反映支持使用多值逻辑的整体网络性能,并估计在面积和功率预算上实现的开销。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
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
A 4-PAM interconnect in network-on-chip for high-throughput and latency-sensitive applications
In this paper, a network-on-chip four-level pulse amplitude modulation (4-PAM) scheme is proposed to be used for communication within the network itself in MPSoCs. A current-mode based 4-PAM transmitter is used to encode data transactions between neighboring routers. Decoding data streams is done by a flash-ADC based receiver using clocked latched type comparators. Additionally, this scheme is implemented on networks utilizing high-radix routers with a local concentration factor of 2 IPs per node to encode data streams injected into the network at the network interface and decode them at the input port of the router. We also discuss the required modifications to the router architecture in the input port buffers and introduce a two-stage allocation method to resolve conflicts of output port requests which is essential to maintain system stability after saturation by utilizing a fair flow control methodology. This results in a reduction in wiring load for each router which is an added value that facilitates the routing stage. The evaluation is extended to reflect the overall network performance supporting the use of multi-valued logic and estimate the overhead of implementation on area and power budgets.
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
期刊最新文献
Body-biasing assisted vmin optimization for 5nm-node multi-Vt FD-SOI 6T-SRAM PDA-HyPAR: Path-diversity-aware hybrid planar adaptive routing algorithm for 3D NoCs A loop structure optimization targeting high-level synthesis of fast number theoretic transform Hybrid-comp: A criticality-aware compressed last-level cache Low power latch based design with smart retiming
×
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