An adaptive greedy flow routing algorithm for performance improvement in software‐defined network

Abstract

AbstractThe software‐defined network (SDN) is an architecture to separate control and data plane. There is a central controller to control the forwarding elements (FEs) under its domain. The controller can receive flows information and requirements to make a decision and impose rules to FEs for forwarding all packets belong to each flow. Routing can be done with considering performance parameters that are needed for each flow. The routing algorithms that support the thresholds of the flows' bandwidth, delay, jitter, and packet loss rate (PLR) are needed in the SDN architecture to optimize the delay, the blocking probability (BP), and the network utilization with an acceptable algorithm running time. In this paper, we proposed the greedy routing scheme in the Floodlight controller and evaluated this algorithm with the utilization, delay, BP, and algorithm time complexity. Extensive simulations are performed, and achieved results confirm the effectiveness of the proposed scheme in comparison with other baseline routing protocols such as random, shortest path (SP), and adaptive flow routing (AFR).