融合智能算法的软件定义网络负载均衡控制

Software Defined Network Load Balancing Control with Intelligent Algorithm

为了使软件定义网络资源分配更加合理,提高网络负载均衡能力,提出融合智能算法的负载均衡控制方法。为了对网络流量进行合理的分配与调度,将系统框架分为信息统计、感知、管理、测量和流表下发5个模块。利用平衡系数来体现实际网络负载情况,当出现负载失衡,则通过调整阈值的方式使负载恢复均衡。根据判决阈值来确定负载的均值水平,并采取智能算法对网络负载失衡状态进行定时的检测,为了减少终端卸载量,利用负载平衡因子对候选终端采取精准搜索。仿真数据显示,由于全面考量了信号强度、接入点负载和终端接收信号强度等因素,平均吞吐量有明显的提高,提升了大约9.2%,而且在平衡网络负载和利用率等方面均表现出良好的性能。

In order to make the software-defined network resource allocation more reasonable and improve the network load-balancing ability,a load-balancing control method integrating an intelligent algorithm is proposed.Firstly,in order to allocate and schedule network traffic reasonably,the system framework was divided into five modules:information statistics,perception,management,measurement,and flow table distribution.Then,the balance coefficient was used to reflect the actual network load.When there was a load imbalance,the load was restored to balance by adjusting the threshold.Finally,the average level of the load was determined according to the decision threshold,and the intelligent algorithm was used to detect the network load imbalance regularly.In order to reduce the amount of terminal unloading,the load balance factor was used to accurately search the candidate terminals.The simulation experimental data show that due to the comprehensive consideration of the factors such as signal strength,access point load,and terminal received signal strength,the average throughput has been significantly improved by about 9.2%,and shows good performance in balancing network load and utilization.