{"title":"Rec-Bench: A tool to create benchmark for reconfigurable computers","authors":"M. Fazlali, A. Zakerolhosseini","doi":"10.1109/SPL.2010.5483012","DOIUrl":null,"url":null,"abstract":"Over the last decade, significant attempts have been made to employ reconfigurable computers to accelerate the computation intensive parts (tasks) of the multimedia applications. However, shareing the reconfigurable fabric between computing processes is an important issue and therefore, several design time and runtime mechanisms have been proposed to tackle this problem. One of the basic requirements in these approaches is a real application workload with which the performance of the proposed approach can be measured. Vast majority of previous researches have been evaluated using random generated task sets or one real hardware implementation per task. In this paper, to convince this weakness, we present Reconfigurable Benchmark (Rec-Bench), a tool to create benchmark suite for reconfigurable computers. The result of applying runtime task scheduling algorithms to the benchmarks in Rec-Bench and random generated tasks justifies the usefulness of using Rec-Bench for the evaluation of runtime resource management algorithm in reconfigurable computers.","PeriodicalId":372692,"journal":{"name":"2010 VI Southern Programmable Logic Conference (SPL)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2010 VI Southern Programmable Logic Conference (SPL)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SPL.2010.5483012","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 3
Abstract
Over the last decade, significant attempts have been made to employ reconfigurable computers to accelerate the computation intensive parts (tasks) of the multimedia applications. However, shareing the reconfigurable fabric between computing processes is an important issue and therefore, several design time and runtime mechanisms have been proposed to tackle this problem. One of the basic requirements in these approaches is a real application workload with which the performance of the proposed approach can be measured. Vast majority of previous researches have been evaluated using random generated task sets or one real hardware implementation per task. In this paper, to convince this weakness, we present Reconfigurable Benchmark (Rec-Bench), a tool to create benchmark suite for reconfigurable computers. The result of applying runtime task scheduling algorithms to the benchmarks in Rec-Bench and random generated tasks justifies the usefulness of using Rec-Bench for the evaluation of runtime resource management algorithm in reconfigurable computers.