Research On Multi-Core And Multi-Thread System Task Scheduling Based On Particle Swarm Optimization Algorithm

Task scheduling in multi-core multi thread system is one of the hot topics in the research of high performance processors.In recent years,many researches have been made on the task scheduling problem of multi-core processors.The aim is to reduce the communication overhead,shorten the length of the task scheduling,and improve the performance of the processor.Most task scheduling problem has been proved to be NP complete problem,and all kinds of scheduling algorithms are given under certain conditions.Task scheduling is from two directions to solve the process of task resource assignment,the two directions are the time and space,a good task scheduling algorithm can greatly improve the comprehensive performance of the multi-core and multi thread system.At present,it is generally believed that the most promising task scheduling technology is heuristic scheduling,such as genetic algorithm,particle swarm algorithm,hoping to find a way to solve such problems in intelligent algorithms.The application of genetic algorithm in task scheduling is relatively complicated and premature convergence,and the convergence rate of the particle swarm optimization algorithm is usually faster than that of the genetic algorithm.Based on the above background,this paper proposes a new algorithm of multi core and multi thread system task scheduling algorithm based on particle swarm optimization algorithm.The model of multi core and multi thread system is established,and the original particle position update mode,fitness function and some parameters can`t be applied to the model.By comparing with the existing multi-core and multi thread system intelligent algorithms(genetic algorithm based on ant colony algorithm and analysis efficiency to obtain the optimal solution,and obtain the optimal solution of the hit rate,an improved particle swarm optimization algorithm is improved to a certain extent.