A Comparative Evaluation of the GPU vs The CPU for Parallelization of Evolutionary Algorithms Through Multiple Independent Runs

摘要

Multiple independent runs of an evolutionary algorithm in parallel are often used to increase the efficiencyof parameter tuning or to speed up optimizations involving inexpensive fitness functions. A GPU platformis commonly adopted in the research community to implement parallelization, and this platform has beenshown to be superior to the traditional CPU platform in many previous studies. However, it is not clearhow efficient the GPU is in comparison with the CPU for the parallelizing multiple independent runs, asthe vast majority of the previous studies focus on parallelization approaches in which the parallel runs aredependent on each other (such as master-slave, coarse-grained or fine-grained approaches). This studytherefore aims to investigate the performance of the GPU in comparison with the CPU in the context ofmultiple independent runs in order to provide insights into which platform is most efficient. This is donethrough a number of experiments that evaluate the efficiency of the GPU versus the CPU in variousscenarios. An analysis of the results shows that the GPU is powerful, but that there are scenarios where theCPU outperforms the GPU. This means that a GPU is not the universally best option for parallelizingmultiple independent runs and that the choice of computation platform therefore should be an informeddecision. To facilitate this decision and improve the efficiency of optimizations involving multipleindependent runs, the paper provides a number of recommendations for when and how to use the GPU.