The Application of a Genetic Algorithm to Global Optimization Problem Solving on Parallel and Distributed Computing Systems

This article presents the results of the adaptation of method of searching the global minimum of multiextremal criterion function of multiple variables with constraints based on genetic algorithm for parallel and distributed computing systems. Two variants of genetic algorithm parallelization are proposed. The reliability and performance of parallel versions of an algorithm, depending on its parameters and the number of nodes in parallel computer system is investigated. The reliability of finding the global minimum of criterion function in a small amount of time with proposed parallel variants of genetic algorithm is represented.