General Methodology for Converting a Sequential Evolutionary Algorithm into Parallel Algorithm with MPI to Water Design Networks

摘要

This paper presents a general methodology for the conversion of the sequential evolutionary algorithms into parallel evolutionary algorithms using MPI. The methodology is showed by applying it to a combinatorial optimization problem well-known as Water Distribution Networks. Also a comparison of two parallelization approaches is done, showing the advantages and disadvantages of each approach. The implementation of this parallel algorithm reduces the necessary time to obtain good solutions for NP-Complete problems as Water Distribution Networks.