Estimating Fine-Grained Mobile Application Energy Use based on Run-Time Software Measured Features

Abstract

Inefficient mobile software kills battery life. Yet, developers lack the tools necessary to detect and solve energy bugs in software. In addition, developers are usually tasked with the creation of software features and triaging existing bugs. This means that most developers do not have the time or resources to research, build, or employ energy debugging tools.We present a new method for predicting software energy consumption to help debug software energy issues. Our approach enables developers to align traces of software behavior with traces of software energy consumption. This allows developers to match run-time energy hot spots to the corresponding execution. We accomplish this by applying state-of-the-art neural network models to predict the time series of energy consumption given a software’s behavior. We compare our time series models to prior state-of-the-art models that only predict total software energy consumption. We found that machine learning based time series models, and LSTM based time series models, can often be more accurate at predicting instantaneous power use and total energy consumption. We also show that the machine learning time series models perform best in terms of learning the shape of the application power usage compared to the other investigated models. This means the time series models are better at modelling the hot spots in online energy consumption than the other tested models.