利用COTS混合精度gpu冗余资源实现高效DWC

2020 20th European Conference on Radiation and Its Effects on Components and Systems (RADECS) Pub Date : 2020-10-01 DOI:10.1109/RADECS50773.2020.9857711

P. M. Basso, F. F. dos Santos, M. Brandalero, P. Rech

{"title":"利用COTS混合精度gpu冗余资源实现高效DWC","authors":"P. M. Basso, F. F. dos Santos, M. Brandalero, P. Rech","doi":"10.1109/RADECS50773.2020.9857711","DOIUrl":null,"url":null,"abstract":"COTS GPUs include redundant resources as tensor-cores and approximate functional-units that can be leveraged for DWC. Using redundant hardware we detect an average of 74% errors with 0.1% time and 24% energy consumption overhead.","PeriodicalId":371838,"journal":{"name":"2020 20th European Conference on Radiation and Its Effects on Components and Systems (RADECS)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"On the Use of Redundant Resources in COTS Mixed-Precision GPUs for Efficient DWC\",\"authors\":\"P. M. Basso, F. F. dos Santos, M. Brandalero, P. Rech\",\"doi\":\"10.1109/RADECS50773.2020.9857711\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"COTS GPUs include redundant resources as tensor-cores and approximate functional-units that can be leveraged for DWC. Using redundant hardware we detect an average of 74% errors with 0.1% time and 24% energy consumption overhead.\",\"PeriodicalId\":371838,\"journal\":{\"name\":\"2020 20th European Conference on Radiation and Its Effects on Components and Systems (RADECS)\",\"volume\":\"7 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2020-10-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2020 20th European Conference on Radiation and Its Effects on Components and Systems (RADECS)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/RADECS50773.2020.9857711\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 20th European Conference on Radiation and Its Effects on Components and Systems (RADECS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/RADECS50773.2020.9857711","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

COTS gpu包括冗余资源，如张量核和近似功能单元，可以用于DWC。使用冗余硬件，我们以0.1%的时间和24%的能耗开销平均检测出74%的错误。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

On the Use of Redundant Resources in COTS Mixed-Precision GPUs for Efficient DWC

COTS GPUs include redundant resources as tensor-cores and approximate functional-units that can be leveraged for DWC. Using redundant hardware we detect an average of 74% errors with 0.1% time and 24% energy consumption overhead.

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

2020 20th European Conference on Radiation and Its Effects on Components and Systems (RADECS)

自引率

0.00%

发文量