DocumentCode
707990
Title
Detecting Display Energy Hotspots in Android Apps
Author
Mian Wan ; Yuchen Jin ; Ding Li ; Halfond, William G. J.
Author_Institution
Univ. of Southern California, Los Angeles, CA, USA
fYear
2015
fDate
13-17 April 2015
Firstpage
1
Lastpage
10
Abstract
Energy consumption of mobile apps has become an important consideration as the underlying devices are constrained by battery capacity. Display represents a significant portion of an app´s energy consumption. However, developers lack techniques to identify the user interfaces in their apps for which energy needs to be improved. In this paper, we present a technique for detecting display energy hotspots - user interfaces of a mobile app whose energy consumption is greater than optimal. Our technique leverages display power modeling and automated display transformation techniques to detect these hotspots and prioritize them for developers. In an evaluation on a set of popular Android apps, our technique was very accurate in both predicting energy consumption and ranking the display energy hotspots. Our approach was also able to detect display energy hotspots in 398 Android market apps, showing its effectiveness and the pervasiveness of the problem. These results indicate that our approach represents a potentially useful technique for helping developers to detect energy related problems and reduce the energy consumption of their mobile apps.
Keywords
Android (operating system); mobile computing; power aware computing; user interfaces; Android market apps; automated display transformation techniques; display energy hotspots detection; display power modeling; energy consumption; mobile app; user interfaces; Color; Energy consumption; Mobile communication; Monitoring; Power demand; Smart phones; User interfaces;
fLanguage
English
Publisher
ieee
Conference_Titel
Software Testing, Verification and Validation (ICST), 2015 IEEE 8th International Conference on
Conference_Location
Graz
Type
conf
DOI
10.1109/ICST.2015.7102585
Filename
7102585
Link To Document