Li Li, Yunhao Bai, Xiaorui Wang, Mai Zheng, Feng Qin
{"title":"Selective checkpointing for minimizing recovery energy and efforts of smartphone apps","authors":"Li Li, Yunhao Bai, Xiaorui Wang, Mai Zheng, Feng Qin","doi":"10.1109/IGCC.2017.8323571","DOIUrl":null,"url":null,"abstract":"Unintended smartphone rebooting and unexpected shutdown (due to reasons like battery run outs, overheating, or automatic app upgrades) is annoying. What can be even worse is that a phone user has to restart, from the very beginning, the apps he or she was using when the phone got rebooted, because all the app states would be lost, especially when the number of apps in use is large. Hence, a recovery service is sorely needed for today's smartphones where apps are becoming increasingly complex. While checkpointing has long been used for desktop and laptop computers, such whole-system state preserving techniques cannot be applied to smartphones directly, due to the constraints of smartphones on energy, delay, and storage space. In this paper, we propose SmartCP, an intelligent checkpointing methodology, in order to reduce the energy required by a smartphone and the amount of efforts required by a user to recover the app states after the smartphone restarts. SmartCP selectively checkpoints the most important apps based on the apps' characteristics and predicted future usage, under the resource constraints of the phone. We propose a novel model that quantitatively analyzes the recovery energy and efforts of each category of smartphone apps and formulate selective checkpointing as a constrained optimization problem. We prototype SmartCP on Android and evaluate it using real-world traces as well as real user feedback. The results show that SmartCP outperforms two alternative app selection schemes by saving 28% more energy and 39% more recovery efforts on average.","PeriodicalId":133239,"journal":{"name":"2017 Eighth International Green and Sustainable Computing Conference (IGSC)","volume":"48 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 Eighth International Green and Sustainable Computing Conference (IGSC)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IGCC.2017.8323571","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
Unintended smartphone rebooting and unexpected shutdown (due to reasons like battery run outs, overheating, or automatic app upgrades) is annoying. What can be even worse is that a phone user has to restart, from the very beginning, the apps he or she was using when the phone got rebooted, because all the app states would be lost, especially when the number of apps in use is large. Hence, a recovery service is sorely needed for today's smartphones where apps are becoming increasingly complex. While checkpointing has long been used for desktop and laptop computers, such whole-system state preserving techniques cannot be applied to smartphones directly, due to the constraints of smartphones on energy, delay, and storage space. In this paper, we propose SmartCP, an intelligent checkpointing methodology, in order to reduce the energy required by a smartphone and the amount of efforts required by a user to recover the app states after the smartphone restarts. SmartCP selectively checkpoints the most important apps based on the apps' characteristics and predicted future usage, under the resource constraints of the phone. We propose a novel model that quantitatively analyzes the recovery energy and efforts of each category of smartphone apps and formulate selective checkpointing as a constrained optimization problem. We prototype SmartCP on Android and evaluate it using real-world traces as well as real user feedback. The results show that SmartCP outperforms two alternative app selection schemes by saving 28% more energy and 39% more recovery efforts on average.