Priyankar Ghosh, Aritra Hazra, Rahul Gonnabhaktula, N. Bhilegaonkar, P. Dasgupta, C. Mandal, Krishna Paul
{"title":"POWER-SIM: An SOC Simulator for Estimating Power Profiles of Mobile Workloads","authors":"Priyankar Ghosh, Aritra Hazra, Rahul Gonnabhaktula, N. Bhilegaonkar, P. Dasgupta, C. Mandal, Krishna Paul","doi":"10.1166/jolpe.2012.1193","DOIUrl":null,"url":null,"abstract":"The power consumption of complex System-On-Chips (SOCs) heavily depends on application profile and usage patterns. Customization of power management strategies based on application as well as user profile have great promise towards minimizing the power wastage. In this paper, we present a methodology for computing the power usage at a higher level of granularity for analyzing the scope of the optimization of the power management strategies in a complex low power SOC with respect to given workload profiles by customizing the boundaries of power domains. The proposed methodology computes the power usage for a given SOC and a given power domain partitioning. We demonstrate the effect of user profiles on the partitioning strategy through a case study.","PeriodicalId":349073,"journal":{"name":"2011 International Symposium on Electronic System Design","volume":"65 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2011-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2011 International Symposium on Electronic System Design","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1166/jolpe.2012.1193","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
The power consumption of complex System-On-Chips (SOCs) heavily depends on application profile and usage patterns. Customization of power management strategies based on application as well as user profile have great promise towards minimizing the power wastage. In this paper, we present a methodology for computing the power usage at a higher level of granularity for analyzing the scope of the optimization of the power management strategies in a complex low power SOC with respect to given workload profiles by customizing the boundaries of power domains. The proposed methodology computes the power usage for a given SOC and a given power domain partitioning. We demonstrate the effect of user profiles on the partitioning strategy through a case study.