Study On Flow Table Optimization Technology In Software Defined Network

Software Defined Network(SDN)is a new type of network architecture,its core idea is that separating the control plane from the data plane of the network,which can operate programmable network through a centralized controller.SDN can manage and control the data flow by means of flow table technology,The routing decision of the controller is sent to the switch by the flow table,and the data flow forward according to matching the flow entry.The flow table is an important part of the data forwarding,Since the flow table resources in the SDN are limited,it will appear flow table overflow and cause network performance degradation.Therefore how to use flow table resources efficiently is worth studying in the SDN.The main work of this article includes the following two aspects:Firstly,this paper analyzes the cause of flow table overflow and observes the impact of flow table overflow through experiments,since flow table management is closely related to routing and the flow table resources are limited,it is necessary to allocate reasonable the flow table resources for the data flow and alleviate the imbalance of flow table utilization.In order to avoid the impact of flow table overflow,this paper proposes a routing mechanism that balancing the bandwidth and the flow table.The experiment results show that the routing mechanism can avoid bottleneck switches,avoid link congestion,and achieve network load balanc e.Secondly,this paper analyzes the influencing factors of the idle_timeout and takes into account the time interval of the data packet and the flow table capacity.In order to use the flow table capacity efficiently,this paper proposes a method for dynamically configuring the idle_timeout on the traffic characteristics,which can use the lognormal distribution of the packet time interval and the flow table capacity to configure an appropriate idle_timeout for the data flow.The experiment results show that the proposed method can improve the utilization of the flow table and improve the performance of the network.