Algorithms, which can process scalable rule sets of virtual routers on a multiprocessor system, have become a hotspot with the improvement of CPU architecture and virtual networks. Therefore, it is important to solve the problems of load balance of processors and stable high-throughput of routers simultaneously. In this paper, the authors build a Dynamic Multiprocessors Selection Model based on evolutionary game and propose a dynamic multiprocessors selection algorithm, called population evolution algorithm, for virtual routers. The population evolution algorithm is designed based on utility function. It needs extra software and hardware to support a centralized utility manager.