摘要： 针对遗传算法应用于服务质量(QoS)组播路由时运算时间长的问题，提出一种基于集群路由器的并行遗传算法，并对集群路由器的体系结构加以改进，使其适用于该并行遗传算法的运行。该算法对种群初始化方法进行改进，使每个从路由器能独立初始化种群，减少主从路由器之间的通信次数，缩短算法收敛的时间。实验结果证明了该算法的有效性。

关键词: 集群路由器, 组播路由, 并行计算, 遗传算法

Abstract: In order to solve the problem that the operation of genetic algorithm needs lots of times when applied to Quality of Service(QoS) multicast routing, a method of parallel genetic algorithm based on cluster route is proposed, and the cluster router architecture is improved to make it applicable for the operation of parallel genetic algorithm. This algorithm improves the initial population, makes the every slave router initializes the population individually, reduces the number of the communication between master and slave routers and shortens the time of the convergence. Experimental results prove the validity of the algorithm.

Key words: cluster router, multicast routing, parallel computing, genetic algorithm

中图分类号: 

• TP393