Abstract

Energy efficiency is an important topic in the area of mobile computing. Developers are often unaware of the impact their choices on data type use and algorithm design have on this non-functional property. Software energy consumption profiling can be utilized to identify the energy behaviour of implemented methods, while pattern mining can be utilized to identify recurring patterns in the methods being run. We present a methodology to combine energy consumption profiling and discriminative pattern mining to identify energy efficiency patterns. In a study of eight sorting algorithms implemented in Java with the data types int, double and Comparable, profiled on the Android platform, we manage to identify significant patterns in the source code of these 24 implementations. The results show that patterns can be identified for both, the data type in use, and for the energy behaviour of efficient or inefficient sorting algorithms, that explain the observed energy profiles.

Original languageEnglish
Title of host publication22nd International Conference on Modeling and Applied Simulation, MAS 2023
EditorsAgostino G. Bruzzone, Fabio De Felice, Francesco Longo, Marina Massei, Adriano O. Solis
PublisherCal-Tek srl
ISBN (Electronic)9788885741928
DOIs
Publication statusPublished - 2023
Event22nd International Conference on Modeling and Applied Simulation, MAS 2023 - Athens, Greece
Duration: 18 Sept 202320 Sept 2023

Publication series

NameProceedings of the International Conference on Modeling and Applied Simulation, MAS
Volume2023-September
ISSN (Print)2724-0037

Conference

Conference22nd International Conference on Modeling and Applied Simulation, MAS 2023
Country/TerritoryGreece
CityAthens
Period18.09.202320.09.2023

Keywords

  • Decision Support Systems
  • Genetic Algorithms
  • Knowledge Based Systems

Fingerprint

Dive into the research topics of 'Identifying Energy Efficiency Patterns in Sorting Algorithms via Abstract Syntax Tree Mining'. Together they form a unique fingerprint.

Cite this