Towards greener software engineering using software analytics
Date
2021-11-12
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Mobiilirakendused, mis ei tühjenda akut, saavad tavaliselt head kasutajahinnangud. Mobiilirakenduste energiatõhusaks muutmiseks on avaldatud mitmeid refaktoreerimis- suuniseid ja tööriistu, mis aitavad rakenduse koodi optimeerida. Neid suuniseid ei saa aga seoses energiatõhususega üldistada, sest kõigi kontekstide kohta ei ole piisavalt energiaga seotud andmeid. Olemasolevad energiatõhususe parandamise tööriistad/profiilid on enamasti prototüübid, mis kohalduvad ainult väikese alamhulga energiaga seotud probleemide suhtes. Lisaks käsitlevad olemasolevad suunised ja tööriistad energiaprobleeme peamiselt a posteriori ehk tagantjärele, kui need on juba lähtekoodi sees. Android rakenduse koodi saab põhijoontes jagada kaheks osaks: kohandatud kood ja korduvkasutatav kood. Kohandatud kood on igal rakendusel ainulaadne. Korduvkasutatav kood hõlmab kolmandate poolte teeke, mis on rakendustesse lisatud arendusprotessi kiirendamiseks. Alustuseks hindame mitmete lähtekoodi halbade lõhnade refaktoreerimiste energiatarbimist Androidi rakendustes. Seejärel teeme empiirilise uuringu Androidi rakendustes kasutatavate kolmandate osapoolte võrguteekide energiamõju kohta. Pakume üldisi kontekstilisi suuniseid, mida võiks rakenduste arendamisel kasutada. Lisaks teeme süstemaatilise kirjanduse ülevaate, et teha kindlaks ja uurida nüüdisaegseid tugitööriistu, mis on rohelise Androidi arendamiseks saadaval. Selle uuringu ja varem läbi viidud katsete põhjal toome esile riistvarapõhiste energiamõõtmiste jäädvustamise ja taasesitamise probleemid. Arendame tugitööriista ARENA, mis võib aidata koguda energiaandmeid ja analüüsida Androidi rakenduste energiatarbimist. Viimasena töötame välja tugitööriista REHAB, et soovitada arendajatele energiatõhusaid kolmanda osapoole võrguteeke
Mobile apps that do not drain the battery usually get good user ratings. To make mobile apps energy efficient many refactoring guidelines and tools are published that help optimize the app code. However, these guidelines cannot be generalized w.r.t energy efficiency, as there is not enough energy-related data for every context. Existing energy enhancement tools/profilers are mostly prototypes applicable to only a small subset of energy-related problems. In addition, the existing guidelines and tools mostly address the energy issues a posteriori, i.e., once they have already been introduced into the code. Android app code can be roughly divided into two parts: the custom code and the reusable code. Custom code is unique to each app. Reusable code includes third-party libraries that are included in apps to speed up the development process. We start by evaluating the energy consumption of various code smell refactorings in native Android apps. Then we conduct an empirical study on the energy impact of third-party network libraries used in Android apps. We provide generalized contextual guidelines that could be used during app development Further, we conduct a systematic literature review to identify and study the current state of the art support tools available to aid green Android development. Based on this study and the experiments we conducted before, we highlight the problems in capturing and reproducing hardware-based energy measurements. We develop the support tool ‘ARENA’ that could help gather energy data and analyze the energy consumption of Android apps. Last, we develop the support tool ‘REHAB’ to recommend energy efficient third-party network libraries to developers.
Mobile apps that do not drain the battery usually get good user ratings. To make mobile apps energy efficient many refactoring guidelines and tools are published that help optimize the app code. However, these guidelines cannot be generalized w.r.t energy efficiency, as there is not enough energy-related data for every context. Existing energy enhancement tools/profilers are mostly prototypes applicable to only a small subset of energy-related problems. In addition, the existing guidelines and tools mostly address the energy issues a posteriori, i.e., once they have already been introduced into the code. Android app code can be roughly divided into two parts: the custom code and the reusable code. Custom code is unique to each app. Reusable code includes third-party libraries that are included in apps to speed up the development process. We start by evaluating the energy consumption of various code smell refactorings in native Android apps. Then we conduct an empirical study on the energy impact of third-party network libraries used in Android apps. We provide generalized contextual guidelines that could be used during app development Further, we conduct a systematic literature review to identify and study the current state of the art support tools available to aid green Android development. Based on this study and the experiments we conducted before, we highlight the problems in capturing and reproducing hardware-based energy measurements. We develop the support tool ‘ARENA’ that could help gather energy data and analyze the energy consumption of Android apps. Last, we develop the support tool ‘REHAB’ to recommend energy efficient third-party network libraries to developers.
Description
Keywords
software development, mobile applications, energy efficiency