Energy-aware load balancing of parallel evolutionary algorithms with heavy fitness functions in heterogeneous CPU-GPU architectures

摘要

By means of the availability of mechanisms such as Dynamic Voltage and Frequency Scaling(DVFS) and heterogeneous architectures including processors with different power consumptionprofiles, it is possible to devise scheduling algorithms that are aware of both runtime and energyconsumption in parallel programs. In this paper, we propose and evaluate amulti-objective (morespecifically, a bi-objective) approach to distribute the workload among the processing cores in agiven heterogeneous parallel CPU-GPU architecture. The aim of this distribution may be eitherto save energy without increasing the running time or to reach a trade-off among time and energyconsumption. The parallel programs considered here aremaster-worker evolutionary algorithmswhere the evaluation of the fitness function for the individuals in the population demands themost part of the computing time. As many useful bioinformatics and data mining applicationsexhibit this kind of parallel profile, the proposed energy-aware approach forworkload schedulingcould be frequently applied.