Performance Improvement of Multicore Processor using Genetic Algorithm

This paper analyzes some practical and technical issues of task scheduling on parallel computing systems which are applied to firm task portions of core-based processors on affine model communications. The need to perform a task fast and efficiently has always been being hot issues in high-performance computing research. The paper proposes a simple and easy method to answer the challenge by implementing a combination method of both back and forth communications from the operation of message passing, genetic algorithms and generating a scheduled task instruction. We build an application program to improve the performance of multicore processor by approaching heuristic analysis of each processor core performance.