Parallel Genetic Algorithm Using Algorithmic Skeleton

Algorithmic skeleton has received attention as an efficient method of parallel programming in recent years. Using the method, the programmer can implement parallel programs easily. In this study, a set of efficient algorithmic skeletons is introduced for use in implementing parallel genetic algorithm (PGA).A performance model is derived for each skeleton that makes the comparison of skeletons possible in order to select the best one for the application. The performance of the selected skeleton can be increased by specifying the virtual topology required by the application.This is a novel approach with no precedent. Nesting of skeletons used here is another novelty of the study which has been employed only in few previous studies.