Datasets for the paper "Search-based energy optimization of some ubiquitous algorithms"

Abstract

Reducing computational energy consumption is of growing importance, particularly at the extremes (i.e. mobile devices and datacentres). Despite the ubiquity of the Java Virtual Machine (JVM), very little work has been done to apply Search Based Software Engineering (SBSE) to minimize the energy consumption of programs that run on it. The paper describes Opacitor, a tool for measuring the energy consumption of JVM programs using a bytecode level model of energy cost. This is combined with generic SBSE approaches to achieve substantial energy savings for three widely-used software components. Multi-Layer Perceptron implementations minimising both energy and error were found, and energy reductions of up to 70% and 39.85% were obtained over the original code for Quicksort and Object-Oriented container classes respectively. This data set includes the raw results from the experiments, generated instances for the quicksort experiments, and test code written for Guava's container classes. Dedicated UnZip software is recommended for accessing the dataset, for example, IZArc.