The Impact of Instant Messaging on the Energy Consumption of Android Devices

Context. One of the primary uses of mobile devices is to send and receive instant messages via messaging apps. However, no evidence is still available about how receiving instant messages impacts the energy consumption of mobile devices.Goal. With this study we aim to empirically assess to what extent the number and distribution of received instant messages impact the energy consumption of Android devices.Method. The subjects of our experiment are WhatsApp and Telegram, two of the most known and used messaging apps. Each run of the experiment lasts 5 minutes and is executed on a Nexus 9 Android device. The independent variables of the experiment are: (i) the frequency of the received messages (i.e., 0, 10, 25, 50 per minute) and (ii) the distribution of messages arrival (i.e., evenly or in bursts). The dependent variable of the experiment is the energy consumption of the Android device in Joules.Results. We confirm that the energy consumption of the Android device tends to be proportional with the number of received messages across both apps. When the number of received messages is fixed, the frequency of their arrival does not significantly impact the energy consumption of the Android device.Conclusions. This study provides evidence that receiving instant messages can largely reduce the battery life of a user’s Android device, even when the number of received messages is relatively low (i.e., 10 messages per minute). Moreover, sending bursts of messages does not lead to significant changes in terms of energy consumption. Developers can use this information to develop new features for their Instant Messaging apps for aggressively bundling messages without the risk of impacting the energy consumption of end users’ devices.

[1]  O. J. Dunn.  Multiple Comparisons among Means , 1961 .

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

[3]  John J. Marciniak,et al.  Encyclopedia of Software Engineering , 1994, Encyclopedia of Software Engineering.

[4]  Luis Cruz,et al.  Catalog of energy patterns for mobile applications , 2019, Empirical Software Engineering.

[5]  N. Cliff.  Dominance statistics: Ordinal analyses to answer ordinal questions. , 1993 .

[6]  Giuseppe Procaccianti,et al.  Assessing the Impact of Service Workers on the Energy Efficiency of Progressive Web Apps , 2017, 2017 IEEE/ACM 4th International Conference on Mobile Software Engineering and Systems (MOBILESoft).

[7]  Luca Ardito,et al.  Methodological Guidelines for Measuring Energy Consumption of Software Applications , 2019, Sci. Program..

[8]  R. Grissom,et al.  Effect sizes for research: A broad practical approach. , 2005 .

[9]  John A. Silvester,et al.  The impact of application signaling traffic on public land mobile networks , 2014, IEEE Communications Magazine.

[10]  W. Kruskal,et al.  Use of Ranks in One-Criterion Variance Analysis , 1952 .

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

[12]  Philip H. Ramsey.  Nonparametric Statistical Methods , 1974, Technometrics.

[13]  Thomas Schäfer,et al.  The Meaningfulness of Effect Sizes in Psychological Research: Differences Between Sub-Disciplines and the Impact of Potential Biases , 2019, Front. Psychol..

[14]  Frank Englert,et al.  Switching Push and Pull: An Energy Efficient Notification Approach , 2014, 2014 IEEE International Conference on Mobile Services.

[15]  Eoin Martino Grua,et al.  A Framework for the Automatic Execution of Measurement-based Experiments on Android Devices , 2020, 2020 35th IEEE/ACM International Conference on Automated Software Engineering Workshops (ASEW).

[16]  Fahim Kawsar,et al.  Energy Efficient Scheduling for Mobile Push Notifications , 2015, EAI Endorsed Trans. Energy Web.

[17]  Jarrett Rosenberg,et al.  Statistical Methods and Measurement , 2008, Guide to Advanced Empirical Software Engineering.

[18]  Ying Zhang,et al.  DelayDroid: Reducing Tail-Time Energy by Refactoring Android Apps , 2015, Internetware.

[19]  Marco Couto,et al.  Energy Refactorings for Android in the Large and in the Wild , 2020, 2020 IEEE 27th International Conference on Software Analysis, Evolution and Reengineering (SANER).

[20]  Magnus C. Ohlsson,et al.  Experimentation in Software Engineering , 2000, The Kluwer International Series in Software Engineering.

[21]  Matti Siekkinen,et al.  Modeling, Profiling, and Debugging the Energy Consumption of Mobile Devices , 2015, ACM Comput. Surv..

[22]  Yingjun Lyu,et al.  Automated Energy Optimization of HTTP Requests for Mobile Applications , 2016, 2016 IEEE/ACM 38th International Conference on Software Engineering (ICSE).

[23]  Dong Yan,et al.  Lightweight energy consumption analysis and prediction for Android applications , 2018, Sci. Comput. Program..

[24]  Abram Hindle,et al.  An exploratory study on assessing the energy impact of logging on Android applications , 2018, Empirical Software Engineering.

[25]  Marco D. Santambrogio,et al.  Morphone.OS: Context-Awareness in Everyday Life , 2013, 2013 Euromicro Conference on Digital System Design.

[26]  Luciano Baresi,et al.  Trends and Challenges for Software Engineering in the Mobile Domain , 2021, IEEE Software.

[27]  Paramvir Bahl,et al.  Fine-grained power modeling for smartphones using system call tracing , 2011, EuroSys '11.