Enhanced scaling and performance of an optical interconnect based on wavelength-division-multiplexed clockwork routing

E. Bravi, D. Cotter
{"title":"Enhanced scaling and performance of an optical interconnect based on wavelength-division-multiplexed clockwork routing","authors":"E. Bravi, D. Cotter","doi":"10.1364/JON.7.000861","DOIUrl":null,"url":null,"abstract":"An optical routing network for applications requiring large numbers of nodes and low latency, such as the interconnection network inside a future high-end supercomputer, is described. The network is a development of an existing architecture based on a combination of clockwork routing and wavelength division multiplexing (WDM-CR). Although using the same underlying photonic technologies as WDM-CR, the new architecture has several advantages, including the following: the network can be scaled to greater numbers of nodes; routing between closely located nodes is more direct, resulting in lower latency and higher overall throughput; arbitration and control mechanisms are simplified; and the need for optical amplification is removed. Results obtained from full discrete-event traffic simulations demonstrate scalability to interconnection networks as large as 4096 nodes in a flat architecture.","PeriodicalId":49154,"journal":{"name":"Journal of Optical Networking","volume":"7 1","pages":"861-875"},"PeriodicalIF":0.0000,"publicationDate":"2008-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1364/JON.7.000861","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Optical Networking","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1364/JON.7.000861","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

An optical routing network for applications requiring large numbers of nodes and low latency, such as the interconnection network inside a future high-end supercomputer, is described. The network is a development of an existing architecture based on a combination of clockwork routing and wavelength division multiplexing (WDM-CR). Although using the same underlying photonic technologies as WDM-CR, the new architecture has several advantages, including the following: the network can be scaled to greater numbers of nodes; routing between closely located nodes is more direct, resulting in lower latency and higher overall throughput; arbitration and control mechanisms are simplified; and the need for optical amplification is removed. Results obtained from full discrete-event traffic simulations demonstrate scalability to interconnection networks as large as 4096 nodes in a flat architecture.
基于波分复用时钟路由的光互连的增强缩放和性能
介绍了一种用于需要大量节点和低延迟的应用的光路由网络,例如未来高端超级计算机内部的互连网络。该网络是基于时钟路由和波分复用(WDM-CR)组合的现有架构的发展。虽然使用与WDM-CR相同的底层光子技术,但新架构具有以下几个优点:网络可以扩展到更多的节点;位置较近的节点之间的路由更直接,导致更低的延迟和更高的总体吞吐量;简化了仲裁和控制机制;并且不需要光学放大。从完全离散事件流量模拟中获得的结果表明,在平面架构中可扩展到4096个节点的互连网络。
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:604180095
Book学术官方微信