Challenges of high-speed channel design on GPU accelerated system

Chun-Lin Liao, Terence Rodrigues, B. Mutnury
{"title":"Challenges of high-speed channel design on GPU accelerated system","authors":"Chun-Lin Liao, Terence Rodrigues, B. Mutnury","doi":"10.1109/EPEPS.2015.7347166","DOIUrl":null,"url":null,"abstract":"High speed servers continue to deploy graphics processing units (GPUs) in them for large volume data computation. Robustness of signal integrity on the high-speed channel is an important factor in these high performance computing (HPC) system designs. In this paper, the simulation and validation / debugging procedure of a 1U / 2 socket rack server that can support up to 4 GPU units was used as exampled to investigate the challenges on such high-speed channel design.","PeriodicalId":130864,"journal":{"name":"2015 IEEE 24th Electrical Performance of Electronic Packaging and Systems (EPEPS)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2015-12-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2015 IEEE 24th Electrical Performance of Electronic Packaging and Systems (EPEPS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/EPEPS.2015.7347166","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1

Abstract

High speed servers continue to deploy graphics processing units (GPUs) in them for large volume data computation. Robustness of signal integrity on the high-speed channel is an important factor in these high performance computing (HPC) system designs. In this paper, the simulation and validation / debugging procedure of a 1U / 2 socket rack server that can support up to 4 GPU units was used as exampled to investigate the challenges on such high-speed channel design.
GPU加速系统中高速通道设计的挑战
高速服务器继续在其中部署图形处理单元(gpu),以进行大量数据计算。高速信道上信号完整性的鲁棒性是高性能计算系统设计的一个重要因素。本文以支持多达4个GPU单元的1U / 2插座机架服务器的仿真和验证/调试过程为例,探讨了高速通道设计面临的挑战。
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