{"title":"别管电池了,我们来玩游戏吧!","authors":"Benedikt Dietrich, S. Chakraborty","doi":"10.1109/ESTIMedia.2014.6962338","DOIUrl":null,"url":null,"abstract":"Today, the battery life of mobile devices like smart-phones stand as a major obstacle between us and our favorite games. Our goal is to solve this problem using customized power management policies tailored for interactive workloads like game applications. In contrast to video processing, power management for gaming applications is a new field of study and has to account for workload changes based on the users input. We have for the first time studied the stability properties of PID controller-based workload prediction schemes - that have been widely used for video applications - and shown that they do not work for games. Instead, new workload models based on time series modeling are better suited for gaming applications. Using them we have developed power management schemes for modern closed-source games using a DLL injection scheme; all previous techniques required instrumenting the game source code to collect execution times of game frames at runtime. Finally, in this paper we also summarize a number of open problems in this area and outline the potential for new research opportunities.","PeriodicalId":265392,"journal":{"name":"2014 IEEE 12th Symposium on Embedded Systems for Real-time Multimedia (ESTIMedia)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2014-11-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":"{\"title\":\"Forget the battery, let's play games!\",\"authors\":\"Benedikt Dietrich, S. Chakraborty\",\"doi\":\"10.1109/ESTIMedia.2014.6962338\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Today, the battery life of mobile devices like smart-phones stand as a major obstacle between us and our favorite games. Our goal is to solve this problem using customized power management policies tailored for interactive workloads like game applications. In contrast to video processing, power management for gaming applications is a new field of study and has to account for workload changes based on the users input. We have for the first time studied the stability properties of PID controller-based workload prediction schemes - that have been widely used for video applications - and shown that they do not work for games. Instead, new workload models based on time series modeling are better suited for gaming applications. Using them we have developed power management schemes for modern closed-source games using a DLL injection scheme; all previous techniques required instrumenting the game source code to collect execution times of game frames at runtime. Finally, in this paper we also summarize a number of open problems in this area and outline the potential for new research opportunities.\",\"PeriodicalId\":265392,\"journal\":{\"name\":\"2014 IEEE 12th Symposium on Embedded Systems for Real-time Multimedia (ESTIMedia)\",\"volume\":\"13 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2014-11-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"14\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2014 IEEE 12th Symposium on Embedded Systems for Real-time Multimedia (ESTIMedia)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ESTIMedia.2014.6962338\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2014 IEEE 12th Symposium on Embedded Systems for Real-time Multimedia (ESTIMedia)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ESTIMedia.2014.6962338","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Today, the battery life of mobile devices like smart-phones stand as a major obstacle between us and our favorite games. Our goal is to solve this problem using customized power management policies tailored for interactive workloads like game applications. In contrast to video processing, power management for gaming applications is a new field of study and has to account for workload changes based on the users input. We have for the first time studied the stability properties of PID controller-based workload prediction schemes - that have been widely used for video applications - and shown that they do not work for games. Instead, new workload models based on time series modeling are better suited for gaming applications. Using them we have developed power management schemes for modern closed-source games using a DLL injection scheme; all previous techniques required instrumenting the game source code to collect execution times of game frames at runtime. Finally, in this paper we also summarize a number of open problems in this area and outline the potential for new research opportunities.