Optimization of Multi-core Task Scheduling based on Improved Particle Swarm Optimization Algorithm
Xiaohui Cheng, Jinqiu Chi · 2019
Embedded heterogeneous multiprocessor systems have a set of processors with different processing capabilities, and task scheduling becomes a key factor in improving system performance. Aiming at the problem of heterogeneous multiprocessor system task scheduling, this paper proposes an improved particle swarm optimization algorithm. By analyzing the computing power of each processor, a scheduling model for independent tasks of heterogeneous multiprocessor systems is established. In the calculation of the fitness value function, this paper rounds up the position value of the particle, so that the particle swarm optimization algorithm can be better applied to the discrete space. At the same time, the strategy of adjusting the inertia weight is used to improve the global convergence speed, overcome the shortcomings of the search performance degradation of the particle in the late iteration, introduce the shrinkage factor and the function factor, improve the learning rule of the particle, and expand the search space of the particle, increasing the diversity of the particle. Compared with traditional algorithms such as genetic algorithm, annealing algorithm, genetic algorithm, ant colony algorithm, local particle swarm optimization algorithm and global particle swarm optimization, it has better convergence and higher computational efficiency. The simulation results show that the improved algorithm proposed in this paper can obtain better scheduling results in a very short time.