{"title":"Dynamic run-time hardware/software scheduling for 3D reconfigurable SoC","authors":"Quang-Hai Khuat, D. Chillet, M. Hübner","doi":"10.1109/ReConFig.2014.7032512","DOIUrl":null,"url":null,"abstract":"In this paper, we present a new online hardware/software (HW/SW) scheduling algorithm for a 3D Reconfigurable SoC platform comprising a multiprocessors layer and a heterogeneous reconfigurable layer. The proposed algorithm decides on the fly whether the tasks will run in SW or HW, at which time, on which processor or in which region of the reconfigurable layer in order to minimize the overall execution time of the application. It evaluates, during runtime, the interest to continue the SW execution of a task or to cancel it for starting a new HW execution of this task from the initial state. By using our algorithm called Hardware/Software algorithm with Software execution Prediction (HSSP), the overall execution time can be reduced by 26 % compared with other existing HW/SW scheduling methods.","PeriodicalId":137331,"journal":{"name":"2014 International Conference on ReConFigurable Computing and FPGAs (ReConFig14)","volume":"12 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2014-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 International Conference on ReConFigurable Computing and FPGAs (ReConFig14)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ReConFig.2014.7032512","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
In this paper, we present a new online hardware/software (HW/SW) scheduling algorithm for a 3D Reconfigurable SoC platform comprising a multiprocessors layer and a heterogeneous reconfigurable layer. The proposed algorithm decides on the fly whether the tasks will run in SW or HW, at which time, on which processor or in which region of the reconfigurable layer in order to minimize the overall execution time of the application. It evaluates, during runtime, the interest to continue the SW execution of a task or to cancel it for starting a new HW execution of this task from the initial state. By using our algorithm called Hardware/Software algorithm with Software execution Prediction (HSSP), the overall execution time can be reduced by 26 % compared with other existing HW/SW scheduling methods.