{"title":"Optimised Global Reduction on QsNetII","authors":"D. Roweth, A. Pittman","doi":"10.1109/CONECT.2005.28","DOIUrl":null,"url":null,"abstract":"In this paper we describe how QsNet/sup II/ supports reduction, a key collective for massively parallel applications. Results from jobs run on a 512-node quad CPU cluster show excellent scaling, with the average time to execute a 2048 process global sum being 22 microsecs.","PeriodicalId":128992,"journal":{"name":"Hot Interconnects","volume":"52 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2005-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Hot Interconnects","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CONECT.2005.28","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
Abstract
In this paper we describe how QsNet/sup II/ supports reduction, a key collective for massively parallel applications. Results from jobs run on a 512-node quad CPU cluster show excellent scaling, with the average time to execute a 2048 process global sum being 22 microsecs.