基于改进蚁群算法的多控制器负载均衡仿真

Simulation of Multi Controller Load Balancing Based on Improved Ant Colony Algorithm

单控制器的灵活性较低,因此通常采用多控制器实现网络均衡,但是多控制器负载均衡普遍存在均衡负载效果不佳的问题,为此提出基于SDN的多控制器负载均衡算法。采集SDN网络的静态负载信息和动态反馈负载信息,判断SDN网络中的过载控制器,并基于优先权的交换机选择策略选取待迁移的交换机;基于博弈论进行SDN多控制器负载均衡博弈,结合改进蚁群算法将上述待迁移的交换机迁移到最优目标控制器中,实现负载均衡;通过改进蚁群算法求解SDN多控制器负载均衡博弈问题,选取最优迁移控制器,将交换机从过载控制器迁移至最优迁移控制器,实现多控制器负载均衡。实验结果表明,所提方法的多控制器负载均衡实现时间最短、效果最好。

In order to address the problem that the flexibility of a single controller is low and the load balancing performance in multi-controller is not performing well,this article put forward a multi-controller load balancing algorithm based on SDN.Firstly,we collected the static load information and dynamic feedback load information of SDN network,thus determining the overload controller in SDN network.Meanwhile,we chose the exchange board to be migrated according to the priority switch selection strategy.Then,we performed SDN multi-controller load balancing game by the game theory.Moreover,we migrated the exchange boards to be migrated to the optimal target controller by the improved ant colony algorithm and thus achieving load balancing.Furthermore,we used the improved ant colony algorithm to solve the problem of SDN multi-controller load balancing game.After that,we selected the optimal migration controller and migrated the switch from the overload controller to the optimal migration controller.Finally,we achieved the multi-controller load balancing.Experimental results show that the proposed method has the shortest implementation time and the best performance for multi-controller load balancing.