模拟退火遗传算法在网络负载均衡中应用研究  

Application of Network Load Balance Based on Simulated Annealing and Genetic Algorithm

作　　者：谭胜兰[1] 

机构地区：[1]东莞理工学院,广东东莞523000

出　　处：《计算机仿真》2011年第12期111-114,共4页Computer Simulation

摘　　要：研究网络负载均衡问题,由于现代网络负载具有高突发性、高实时等特点,针对网络资源利用率和流量控制问题,单纯的采用模拟退火或遗传算法无法适合其变化规律,导致网络资源利用率低,网络拥塞现象严重。为了提高网络资源利用率,使网络负载均衡更加合理,提出一种与模拟退火和遗传算法相融合的网络负载均衡方法。首先利用遗传算法对网络负载均衡问题进行全局搜索,使问题的解迅速处于全局最优区域附近,然后采用模拟退火算法在全局最优区域附近进行进一步局部寻优,找到网络负载均衡最优解。仿真结果表明,改善了全局搜索速度,提高了网络资源利用率,明显改善网络负载不均衡的状况。Network load balance problem is researched.Because modern network loads have the characteristics of high burstiness and high real-timing,the simple genetic algorithm or simulated annealing cannot fit the change rules,and therefore network resource utilization rate is low and the network has serious congestions.In order to improve the network resource utilization rate,make the network load balance more reasonable,the paper put forward a network load balance method based on simulated annealing and genetic algorithm.Firstly,genetic algorithm was use to sovle the network load balance problems on global search,which quickly made the solution in global optimal regions nearby.Then,the simulated annealing algorithm was used in searching near the global optimal area and found the network load balance optimal solutions.The simulation results show that the method has the advantages of simulated annealing and genetic algorithm,overcomes their shortcomings,and improves the network resource utilization and network load imbalanced status.

关 键 词：网络负载 模拟退火算法 遗传算法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]