基于模糊满意度的软件定义网络交换机迁移策略

针对软件定义网络中多控制器间负载均衡的迁移代价和迁移效率问题,提出一种基于模糊满意度的交换机迁移策略。首先,构建均衡判断矩阵,监测控制器负载状态,且提出交换机组选择度,划分交换机组以选取迁移交换机;其次,考虑迁移代价和负载均衡率,构建模糊满意度迁移竞争模型,且提出改进蚁群算法优化求解,选择最佳迁入控制器;最后,将迁移交换机迁往迁入控制器,实现交换机的快速迁移。实验结果表明,与现有交换机迁移策略相比,所提方法在保证较高负载均衡率的同时,进一步优化了网络性能,迁移代价平均减少约26.8%,控制器平均响应时间缩短约0.41 s,改善了交换机迁移过程。

Switch migration strategy based on fuzzy satisfaction in software defined network

To solve the problem of migration cost and efficiency of load balancing among multiple controllers in software defined networking, this paper proposed a switch migration strategy based on fuzzy satisfaction. Firstly, it constructed the balanced judgment matrix to monitor the load state of controllers, and divided switch groups by the proposed switch unit selection degree to select migration switches. Then, it constructed the migration competition model based on fuzzy satisfaction by considering the migration cost and load balance rate, and proposed an improved ant colony algorithm to select the best migration controller. Finally, it migrated migration switches to the migration controller to realize the rapid migration of switches. Experimental results show that compared with the existing switch migration strategies, this strategy not only ensures the high load balancing rate, but also further optimizes the network performance. In this strategy, the average migration cost decreases by about 26.8% and the average response time reduces by about 0.41 s, which improves the switch migration process.