Research On Traffic Identification And Scheduling Towards SDN

With the explosive growth of Internet services,different types of traffic in the network are becoming more and more abundant,which makes it more difficult to manage and maintain the network.In order to ensure the end-to-end services of different flow,it is necessary to provide effective quality-of-service assurance strategies to meet the needs of various services for the underlying network resources.Under the traditional network architecture,the distributed network equipment deployment scheme results in limited global view,and the routing scheme of hop by hop decision has insufficient control over the traffic,resulting in low network utilization.The proposal of Software Defined Network(SDN)provides a new approach to solve this problem.Through the architecture advantage of turn-by-turn control separation,SDN can monitor the global Network topology and link state in real time,and realize fine traffic scheduling.This thesis uses the efficient management of SDN to the global network to study the identification and scheduling of different traffic in the SDN network environment based on the OpenFlow protocol.The research content includes: 1)According to the characteristics of real-time extracted packets,a real-time SDN traffic identification method based on QoS constraints is designed.2)This thesis studies the method of measuring network state parameters by OpenFlow protocol in SDN environment,and measures link performance parameters such as bandwidth,delay,jitter,packet loss rate by using the communication mechanism between switch and controller.3)Designing an SDN real-time traffic scheduling strategy based on genetic algorithm optimization,The fitness function is designed with the current traffic identification and link real-time state as the factor and ensure that the demand of different services for the underlying network resources is fully considered in the process of path planning,which effectively reduce the network load caused by the chain congestion,and realize the efficient utilization of network resources.Based on the research of traffic identification and scheduling,this thesis uses the open source controller OpenDaylight for secondary development,and implements a traffic identification and scheduling system oriented to SDN.In the control plane,the REST API is developed to extend the functions of packet parsing and real-time state measurement of links to complete real-time identification and routing optimization of business flows.By setting up the simulation topology in Mininet for testing,By building a simulation topology in Mininet for testing,the system's optimal scheduling effect on different business flows is verified.