摘要
并行计算能够有效地缩减求解大规模问题的时间。文中在介绍了粒子群算法(Particle Swarm Optimization algo-rithm)的基础上,对PSO算法的同步异步模型进行分析,给出了并行环境下的同步异步PSO算法。该并行算法在联想深腾1800大型计算机上测试。实验证明PSO算法具有较高的并行性,并行算法明显提高了求解的速度。
Parallel computing can curtail the time of solving large-scale problems effectively. Under introducing the biologic character of particle swarm optimization algorithm (PSO) and analyzing synchronous and asynchronous model, presented the synchronous and asynchronous algorithm of particle swarm optimization under parallel circumstance. The parallel algorithm was tested in the cluster of Lenovo Shenton 1800. The experimentations proved that PSO algorithm had higher parallelism, and the parallel arithmetic improved the efficiency obviously.
出处
《计算机技术与发展》
2009年第3期123-126,共4页
Computer Technology and Development
基金
中央级公益性科研院基本科研业务费专项资金(HKY-JBYW-2007-22)
关键词
PSO算法
并行计算
并行PSO算法
同步
异步
PSO algorithm
parallel computing
parallel algorithm of PSO
synchronous
asynchronous