Research On Load Balancing Routing Algorithm In Software Defined Network

In recent years,with the rapid development of Internet technology and emerging new application scenarios,users have higher and higher requirements for network performance and business applications.The current networks are facing many problems,such as highly dynamic network load and network traffic routing problems.Software defined network(SDN)has the advantages of separation of forwarding layer and control layer and network programmability,its emergence provides a new idea for solving these problems.Based on SDN technology,the combination of network load balancing and traffic routing is studied and designed to improve QoS(quality of service)performance in multimedia network.Firstly,a load balancing routing algorithm is proposed based on QoS priority.Considering the needs of different users,the interference between different services and the imbalance of real-time link load,the priority of service flow is graded more finely by using the two-level classification method.In order to balance the network traffic and the network load,the cost of network traffic routing is optimized,and the differentiated routing of different priority traffic flows is realized.The simulation results show that compared with the existing algorithms,the proposed load balancing routing algorithm based on QoS priority can improve the QoS performance of traffic flows,and allocate network resources and network load balancing between different traffic flows effectively.Secondly,as the continuous expansion of network services,the congestion problem of existing networks becomes more and more serious.Aiming at this problem,a load balancing routing algorithm based on congestion prediction is proposed.The algorithm uses SDN technology to monitor the real-time traffic transmission speed and bandwidth occupation in each link,predict the remaining bandwidth of the link at the next time,introduce bandwidth threshold parameters into the link,predict the congestion probability of the link in real time,and adjust the traffic allocation and dynamically adjust the transmission strategy according to the prediction results and the constraints of network flow in network routing.Simulation results show that compared with existing algorithms,the proposed load balancing routing algorithm based on congestion prediction can reduce link load,network packet loss rate and average delay,and improve network throughput.Finally,since the elephant flow is one of the main factors in leading to network congestion,the load balancing problem of elephant flow is studied and a load balancing routing algorithm based on elephant flow is proposed.In this algorithm,the elephant flow is detected by adjusting the threshold dynamically and adaptively,and the polling cycle can be adjusted dynamically and adaptively to reduce the overhead of the controller.For the candidate path of elephant flow rerouting,the algorithm first generates the selection probability of each link based on the transmission rate and remaining bandwidth of the link,and then selects the rerouting path combined with the random probability selection method.Simulation results show that compared with traditional methods,the proposed load balancing routing algorithm based on elephant flow has better performance in improving network performance and balancing network load.