{"title":"Thermal management via task scheduling for 3D NoC based multi-processor","authors":"Han Wang, Yuzhuo Fu, Ting Liu, Jiafang Wang","doi":"10.1109/SOCDC.2010.5682875","DOIUrl":null,"url":null,"abstract":"3D IC technology drives Network-On-Chip (NoC) design on towards 3D trend and relevant multi-core system further development. Thermal issue becomes more critical because of the increasing power density caused by 3D stack. OS-level task scheduling is an effective method to improve on-chip temperature condition. In this paper we propose a thermal management method using task scheduling to limit chip temperature under required constraints as well as consider performance degradation caused by moving task to a core far away from its data. A temperature controller is implemented in our simulator to determine temperature management actions. The result shows our algorithm has lower performance loss with the same temperature constraint compared to coldest and random scheduling.","PeriodicalId":380183,"journal":{"name":"2010 International SoC Design Conference","volume":"22 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2010-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"21","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2010 International SoC Design Conference","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SOCDC.2010.5682875","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 21
Abstract
3D IC technology drives Network-On-Chip (NoC) design on towards 3D trend and relevant multi-core system further development. Thermal issue becomes more critical because of the increasing power density caused by 3D stack. OS-level task scheduling is an effective method to improve on-chip temperature condition. In this paper we propose a thermal management method using task scheduling to limit chip temperature under required constraints as well as consider performance degradation caused by moving task to a core far away from its data. A temperature controller is implemented in our simulator to determine temperature management actions. The result shows our algorithm has lower performance loss with the same temperature constraint compared to coldest and random scheduling.