Parallel evolutionary computation to solve combinatorial optimization problem

Grid computing and Peer-to-Peer (P2P) computing have become a real alternatives to traditional supercomputing for the development of applications that harness massive computational resources especially applications aiming to solve the NP-Complete problem such as metaheuristics. One the promising methods is parallel programming. This concept is a powerful and necessary way to reduce the computation time of metaheuristics and/or improve the quality of the provided solutions. Different models have been proposed to exploit the parallelism of metaheuristics. These models have been and are still being largely experimented on a wide range of metaheuristics and applied to a large variety of problems in different areas. In this paper. we have proposed a hybrid parallel implementation of Ant Colony Optimization (ACO) and Genetic Algorithm (AG) by using the MPJ Express approach.