Research on Traffic Scheduling Algorithm Based on Link Criticality in SDN Environment

Software-Defined Networks (SDN) traffic scheduling algorithms are critical for improving network performance and resource utilization, achieving energy efficiency, and load balancing. In high-traffic environments, the traffic of some links in the network approaches or exceeds its capacity limit, triggering link congestion and possible link failures. For this reason, this paper proposes a link criticality-based traffic scheduling algorithm, TSA-LC, which implements a rerouting strategy mainly for large flows on congested links in the network. First, the KSP algorithm is used to filter the feasible shortest path set for large flows on congested links. Then, link criticality metrics related to parameters such as link utilization and median centrality are designed to calculate the criticality of all paths in the path set. Finally, these paths are evaluated comprehensively by the fitness function, and the path with the highest fitness is selected as the rerouting path. The algorithm effectively reduces the risk of link congestion and failure and improves the overall performance of the network.