Tales from the Code #1: The Effective Impact of Code Refactorings on Software Energy Consumption

Software energy consumption is a performance related non-functional requirement that complicates building software on mobile devices today. Energy hogging applications are a liability to both the end-user and software developer. Measuring software energy consumption is non-trivial, requiring both equipment and expertise, yet many researchers have found that software energy consumption can be modelled. Prior works have hinted that with more energy measurement data one can make more accurate energy models but this data was expensive to extract because it required energy measurement of running test cases (rare) or time consuming manually written tests. We address these concerns by automatically generating test cases to drive applications undergoing energy measurement. Automatic test generation allows a model to be continuously improved in a model building process whereby applications are extracted, tests are generated, energy is measured and combined with instrumentation to train a grander big-data model of software energy consumption. This continuous process has allowed the authors to generate and extract measurements from hundreds of applications in order to build accurate energy models capable of predicting the energy consumption of applications without end-user energy measurement. We clearly show that models built from more applications reduce energy modelling error.

Download Full-text

Did I make a mistake? Finding the impact of code change on energy regression

10.7287/peerj.preprints.2419v3 ◽

2017 ◽

Author(s):

Shaiful Alam Chowdhury ◽

Stephanie Gil ◽

Stephen Romansky ◽

Abram Hindle

Keyword(s):

Energy Consumption ◽

Model Building ◽

Continuous Process ◽

Measurement Data ◽

Energy Measurement ◽

Test Cases ◽

End User ◽

Energy Models ◽

Software Energy Consumption ◽

The Impact

Software energy consumption is a performance related non-functional requirement that complicates building software on mobile devices today. Energy hogging applications are a liability to both the end-user and software developer. Measuring software energy consumption is non-trivial, requiring both equipment and expertise, yet many researchers have found that software energy consumption can be modelled. Prior works have hinted that with more energy measurement data one can make more accurate energy models but this data was expensive to extract because it required energy measurement of running test cases (rare) or time consuming manually written tests. We address these concerns by automatically generating test cases to drive applications undergoing energy measurement. Automatic test generation allows a model to be continuously improved in a model building process whereby applications are extracted, tests are generated, energy is measured and combined with instrumentation to train a grander big-data model of software energy consumption. This continuous process has allowed the authors to generate and extract measurements from hundreds of applications in order to build accurate energy models capable of predicting the energy consumption of applications without end-user energy measurement. We clearly show that models built from more applications reduce energy modelling error.

Download Full-text

Experiments on Model-Based Software Energy Consumption Analysis Involving Sorting Algorithms

Revista de Informática Teórica e Aplicada ◽

10.22456/2175-2745.98904 ◽

2020 ◽

Vol 27 (3) ◽

pp. 72-83

Author(s):

Danilo Silva Alves ◽

Oseias Ayres Ferreira ◽

Lucio Mauro Duarte ◽

Davi Silva ◽

Paulo Henrique Maia

Keyword(s):

Energy Consumption ◽

Energy Efficient ◽

Measurement Tool ◽

Analysis Tool ◽

Related Information ◽

Energy Consumption Analysis ◽

Sorting Algorithms ◽

Software Energy Consumption ◽

Energy Information ◽

Efficient Software

Although energy has become an important aspect in software development, little support exists for creating energy-efficient programs. One reason for that is the lack of abstractions and tools to enable the analysis of relevant properties involving energy consumption. This paper presents the results of some experiments involving the gathering, modelling, and analysis of energy-related information, in particular, the costs of executing certain parts of a software. We combine some existing free and open-source tools to carry out the experiments, extending one of them to handle energy information. Our experiments consider a comparison of energy consumption of Java implementations of the Bubble Sort, Insertion Sort and Selection Sort algorithms using different data structures. We show how to combine an energy measurement tool and a model analysis tool to carry such a comparison. Based on this support and on our experiments, we believe this is a first step to allow developers to start creating more energy-efficient software.

Download Full-text