Demming regressive multiobjective dragonfly optimized controller placement in SDN environment

Abstract

Wireless networks are of high significance in current telecommunication systems, and in order to improve these systems, Software-Defined Networks (SDN)is used to centrally monitor and control the whole network with the help of a controller. The design of an SDN-based network is needed to identify the optimal amount of controllers for improving performance of network. Controller placement problem is determined on propagation latency but it failed to consider the load balancing, fault tolerance, bandwidth consumption and data transmission rate. Plan to develop a novel technique called Demming Regressive Multiobjective Dragonfly Optimized Controller Placement (DRMDOCP) for an optimal number of controller placement for enhancing network performance during different topologies. By applying DRMDOCP, the optimum number of controllers are selected and placed into the network to improve the overall networks performance. Therefore, delay minimization-based controller placement strategy is extremely preferred. The simulation of DRMDOCP technique and existing methods is conducted using network topology dataset and different performance metrics. The simulation results demonstrate that the DRMDOCP increases the packet delivery, throughput and reduces the average latency, packet drop when compared to state-of-the-art method.