Research On Multi-priority Data Flow Scheduling Mechanism For SDN

Software defined network(SDN)is programmable through the software modules in the controller to control the underlying hardware and realize the on-demand allocation of network resources.In order to guarantee the quality of service(Qo S)of new services such as video on demand and online video conference in SDN,it is necessary to study the scheduling strategy of multi priority data flow in SDN environment and analyze the performance of network models under different scheduling strategies.The main contributions of this article are as follows.1.In order to reduce the end-to-end delay of SDN network system,a system model based on SDN multi priority data flow queue scheduling and load balancing transmission mechanism is constructed,which guarantees the quality of service of multi priority data flow with different real-time requirements,and increases the stability and reliability of data transmission in the network.2.By using stochastic network algorithm,the end-to-end delay performance of the system model with GPS,non preemptive and preemptive scheduling algorithms is analyzed and compared.The simulation results show that the delay performance of the system model using GPS scheduling in SDN network is higher than that of the system model under non preemptive scheduling and preemptive scheduling,which can more effectively guarantee the service quality of different priority data flows,especially the high priority data flows.3.In order to deal with the frequent large traffic burst phenomenon in the actual network,a hybrid adaptive scheduling algorithm is proposed for the system model proposed in Chapter 3.The end-to-end delay and backlog performance of the system model under this algorithm are analyzed,and compared with the delay and backlog performance of the system model under GPS scheduling in chapter 3.The simulation results show that the delay performance of the system model under this scheduling algorithm is higher than that under GPS scheduling,and its backlog performance is better.Therefore,the system model based on the hybrid adaptive scheduling algorithm is more suitable for the actual network.4.The system model based on GPS and hybrid adaptive scheduling algorithm is tested to ensure the correctness of theoretical analysis.This paper mainly studies the end-to-end delay performance of SDN network system.Firstly,the system model of multi priority data flow queue scheduling and load balancing transmission mechanism is built to improve the service ability of SDN network to data flow.Then,the stochastic network algorithm is used to analyze the system model under different scheduling algorithms.Finally,the network experiment is used to verify the correctness of the theoretical derivation.