Performance Investigation of Parallel Immune Plasma Algorithm

Abstract

The increasing power and common usage of the shared memory based systems gained the researchers' attentions to the development of parallel algorithms or parallelization of the existing algorithms for utilizing the computational capabilities of these systems. In this study, Immune Plasma algorithm for short IP Algorithm or IPA that models the steps of a medical method called convalescent plasma or immune plasma treatment was parallelized and solving capabilities of the parallel IPA was investigated by using a set of numerical benchmark problems. The results of the experimental studies showed that parallelization of IPA contributes to the convergence performance and qualities of the final solutions while decreasing the execution time compared to the serial implementation of the same algorithm. © 2021 IEEE.