Mining API Interactions to Analyze Software Revisions for the Evolution of Energy Consumption (2103.11377v1)

Abstract: With the widespread use and adoption of mobile platforms like Android a new software quality concern has emerged -- energy consumption. However, developing energy-efficient software and applications requires knowledge and likewise proper tooling to support mobile developers. To this aim, we present an approach to examine the energy evolution of software revisions based on their API interactions. The approach stems from the assumption that the utilization of an API has direct implications on the energy being consumed during runtime. Based on an empirical evaluation, we show initial results that API interactions serve as a flexible, lightweight, and effective way to compare software revisions regarding their energy evolution. Given our initial results we envision that in future using our approach mobile developers will be able to gain insights on the energy implications of changes in source code in the course of the software development life-cycle.