How Android App Developers Manage Power Consumption? - An Empirical Study by Mining Power Management Commits

As Android platform becomes more and more popular, a large amount of Android applications have been developed. When developers design and implement Android applications, power consumption management is an important factor to consider since it affects the usability of the applications. Thus, it is important to help developers adopt proper strategies to manage power consumption. Interestingly, today, there is a large number of Android application repositories made publicly available in sites such as GitHub. These repositories can be mined to help crystalize common power management activities that developers do. These in turn can be used to help other developers to perform similar tasks to improve their own Android applications.In this paper, we present an empirical study of power management commits in Android applications. Our study extends that of Moura et al. who perform an empirical studyon energy aware commits; however they do not focus on Android applications and only a few of the commits that they study come from Android applications. Android applications are often different from other applications (e.g., those running on a server) due to the issue of limited battery life and the use of specialized APIs. As subjects of our empirical study, we obtain a list of open source Android applications from F-Droid and crawl their commits from Github. We get 468 power management commits after we filter the commits using a set of keywords and by performing manual analysis. These 468 power management commits are from 154 different Android applications and belong to 15 different application categories. Furthermore, we use open card sort to categorize these power management commits and we obtain 6 groups which correspond to different power management activities. Our study also reveals that for different kinds of Android application (e.g., Games, Connectivity, Navigation, etc.), the dominant power management activities differ.For example, the percentageof power management commits belonging to Power Adaptation activity is larger for Navigation applications than those belonging to other categories.

[1]  Michael W. Godfrey,et al.  The MSR Cookbook: Mining a decade of research , 2013, 2013 10th Working Conference on Mining Software Repositories (MSR).

[2]  Felipe Ebert,et al.  Mining Energy-Aware Commits , 2015, 2015 IEEE/ACM 12th Working Conference on Mining Software Repositories.

[3]  Alan Messer,et al.  Adaptive offloading for pervasive computing , 2004, IEEE Pervasive Computing.

[4]  W.J. Kaiser,et al.  The low power energy aware processing (LEAP) embedded networked sensor system , 2006, 2006 5th International Conference on Information Processing in Sensor Networks.

[5]  Sara Sartoli,et al.  Poster: Reasoning Based on Imperfect Context Data in Adaptive Security , 2015, 2015 IEEE/ACM 37th IEEE International Conference on Software Engineering.

[6]  Ding Li,et al.  An Empirical Study of the Energy Consumption of Android Applications , 2014, 2014 IEEE International Conference on Software Maintenance and Evolution.

[7]  Lori L. Pollock,et al.  SEEDS: a software engineer's energy-optimization decision support framework , 2014, ICSE.

[8]  William G. J. Halfond,et al.  How Does Code Obfuscation Impact Energy Usage? , 2014, 2014 IEEE International Conference on Software Maintenance and Evolution.

[9]  Paramvir Bahl,et al.  The Case for VM-Based Cloudlets in Mobile Computing , 2009, IEEE Pervasive Computing.

[10]  Ming Zhang,et al.  Where is the energy spent inside my app?: fine grained energy accounting on smartphones with Eprof , 2012, EuroSys '12.

[11]  Ramesh Govindan,et al.  Estimating mobile application energy consumption using program analysis , 2013, 2013 35th International Conference on Software Engineering (ICSE).

[12]  Philip J. Guo,et al.  Characterizing and predicting which bugs get reopened , 2012, 2012 34th International Conference on Software Engineering (ICSE).

[13]  Ming Zhang,et al.  Bootstrapping energy debugging on smartphones: a first look at energy bugs in mobile devices , 2011, HotNets-X.

[14]  Timo Smura,et al.  Energy efficiency of mobile handsets: Measuring user attitudes and behavior , 2012, Telematics Informatics.

[15]  Lori L. Pollock,et al.  Initial explorations on design pattern energy usage , 2012, 2012 First International Workshop on Green and Sustainable Software (GREENS).

[16]  Byung-Gon Chun,et al.  CloneCloud: elastic execution between mobile device and cloud , 2011, EuroSys '11.

[17]  Luis Ceze,et al.  Characterizing the Performance and Energy Efficiency of Lock-Free Data Structures , 2011, 2011 15th Workshop on Interaction between Compilers and Computer Architectures.

[18]  Chantal Ykman-Couvreur,et al.  Incorporating energy efficient data structures into modular software implementations for internet-based embedded systems , 2002, WOSP '02.

[19]  Klara Nahrstedt,et al.  Energy-efficient soft real-time CPU scheduling for mobile multimedia systems , 2003, SOSP '03.

[20]  Margaret Martonosi,et al.  Wattch: a framework for architectural-level power analysis and optimizations , 2000, Proceedings of 27th International Symposium on Computer Architecture (IEEE Cat. No.RS00201).

[21]  Uwe Aßmann,et al.  Energy Consumption and Efficiency in Mobile Applications: A User Feedback Study , 2013, 2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical and Social Computing.

[22]  Abram Hindle,et al.  Green mining: a methodology of relating software change and configuration to power consumption , 2013, Empirical Software Engineering.

[23]  Eli Tilevich,et al.  Reducing the Energy Consumption of Mobile Applications Behind the Scenes , 2013, 2013 IEEE International Conference on Software Maintenance.

[24]  Gabriele Bavota,et al.  Mining energy-greedy API usage patterns in Android apps: an empirical study , 2014, MSR 2014.

[25]  Abram Hindle.  Green mining: Investigating power consumption across versions , 2012, 2012 34th International Conference on Software Engineering (ICSE).

[26]  Ramesh Govindan,et al.  Calculating source line level energy information for Android applications , 2013, ISSTA.

[27]  Gustavo Pinto,et al.  Mining questions about software energy consumption , 2014, MSR 2014.

[28]  David Lo,et al.  How practitioners perceive the relevance of software engineering research , 2015, ESEC/SIGSOFT FSE.

[29]  Ting Cao,et al.  The Yin and Yang of power and performance for asymmetric hardware and managed software , 2012, 2012 39th Annual International Symposium on Computer Architecture (ISCA).

[30]  Sven Apel,et al.  Views on Internal and External Validity in Empirical Software Engineering , 2015, 2015 IEEE/ACM 37th IEEE International Conference on Software Engineering.

[31]  Frank Bellosa,et al.  Cooperative I / O-- A Novel I / O Semantics for Energy-Aware Applications , 2003 .

[32]  Jared Smith,et al.  A Dataset of Open-Source Android Applications , 2015, 2015 IEEE/ACM 12th Working Conference on Mining Software Repositories.

[33]  Alan Messer,et al.  Towards a distributed platform for resource-constrained devices , 2002, Proceedings 22nd International Conference on Distributed Computing Systems.

[34]  Ding Li,et al.  An investigation into energy-saving programming practices for Android smartphone app development , 2014, GREENS 2014.

[35]  Ying Zhang,et al.  Refactoring android Java code for on-demand computation offloading , 2012, OOPSLA '12.

[36]  Ramesh Govindan,et al.  Estimating Android applications' CPU energy usage via bytecode profiling , 2012, 2012 First International Workshop on Green and Sustainable Software (GREENS).