OTORNoC:用于1000核系统的片上环网光树

2017 7th International Symposium on Embedded Computing and System Design (ISED) Pub Date : 2017-12-01 DOI:10.1109/ISED.2017.8303940

Soumyajit Poddar, Suraj, Amit Kumar Yadav, H. Rahaman

{"title":"OTORNoC:用于1000核系统的片上环网光树","authors":"Soumyajit Poddar, Suraj, Amit Kumar Yadav, H. Rahaman","doi":"10.1109/ISED.2017.8303940","DOIUrl":null,"url":null,"abstract":"Optical Networks on Chip are an emerging communication platform for high performance multi core systems on chip. In this work a novel thousand core topology is proposed that increases optical link utilization and scales up bandwidth to enable single chip TeraFlops performance. Performance benefits of about 1.8× and 3× reduction in optical power is obtained over the existing state of the art multicore chips.","PeriodicalId":147019,"journal":{"name":"2017 7th International Symposium on Embedded Computing and System Design (ISED)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"OTORNoC: Optical tree of rings network on chip for 1000 core systems\",\"authors\":\"Soumyajit Poddar, Suraj, Amit Kumar Yadav, H. Rahaman\",\"doi\":\"10.1109/ISED.2017.8303940\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Optical Networks on Chip are an emerging communication platform for high performance multi core systems on chip. In this work a novel thousand core topology is proposed that increases optical link utilization and scales up bandwidth to enable single chip TeraFlops performance. Performance benefits of about 1.8× and 3× reduction in optical power is obtained over the existing state of the art multicore chips.\",\"PeriodicalId\":147019,\"journal\":{\"name\":\"2017 7th International Symposium on Embedded Computing and System Design (ISED)\",\"volume\":\"4 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2017-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2017 7th International Symposium on Embedded Computing and System Design (ISED)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ISED.2017.8303940\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 7th International Symposium on Embedded Computing and System Design (ISED)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ISED.2017.8303940","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

片上光网络是面向高性能多核片上系统的新兴通信平台。在这项工作中，提出了一种新的千核拓扑结构，可以提高光链路利用率并扩展带宽，从而实现单芯片TeraFlops性能。与现有的多核芯片相比，光功率降低了1.8倍和3倍的性能优势。

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

微信好友朋友圈 QQ好友复制链接

本刊更多论文

OTORNoC: Optical tree of rings network on chip for 1000 core systems

Optical Networks on Chip are an emerging communication platform for high performance multi core systems on chip. In this work a novel thousand core topology is proposed that increases optical link utilization and scales up bandwidth to enable single chip TeraFlops performance. Performance benefits of about 1.8× and 3× reduction in optical power is obtained over the existing state of the art multicore chips.

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

2017 7th International Symposium on Embedded Computing and System Design (ISED)

自引率

0.00%

发文量