Multi-Core Real-Time Scheduling Algorithm Based on Particle Swarm optimization Algorithm

Abstract

Priority-based scheduling algorithms have a wide range of applications in real-time systems. In today’s commonly used task scheduling algorithms, only the shortest scheduling time is used as the only criterion, while ignoring the importance of task priority. Task allocation among multiple cores is also difficult to balance. At this time, traditional priority scheduling shows great limitations. In order to build a preemptive priority scheduling algorithm for load balancing among multiple cores, this paper first studies the scheduling principles of particle swarm algorithm and annealing algorithm among CPU nodes, and simulates the traditional scheduling algorithm that may appear before the optimal solution for scheduling is obtained. Problem, and then extract the mathematical model of operating system scheduling. Based on the particle swarm algorithm and combined with the scheduling advantages of other heuristic algorithms, an optimized preemptive priority scheduling algorithm based on particle swarm algorithm is proposed, which completes multitasking among multiple cores. Based on the priority of the process, the process is assigned to a more suitable processor core, which improves the efficiency of priority scheduling of heterogeneous multi-core operating systems. Finally, the effectiveness of the algorithm is verified by simulation experiments.