Ant Colony Optimization-based Dynamic Routing in Software Defined Networks

Abstract

Traditional networks are unable to serve the purpose of service providers because of the tight coupling of control and data plane. Software-Defined Network (SDN) is a new and novel way of network management that breaks the vertical integration of traditional networks. SDN decouples the control and data plane where the control plane is responsible for making decisions, and the data plane forwards the packet as the decisions taken by the control plane. Routing in the traditional network is quite challenging as it causes the inefficient utilization of resources when the packets are flooded across the network. SDN offers an efficient routing mechanism as compared to traditional networks because of its decoupled architecture. In this paper, we proposed dynamic routing to compute the path between source and destination using Ant Colony Optimization implemented in SDN. Mininet emulator with POX as as remote controller is used to implement the proposed approach. Performance differentials are analyzed using different scenarios in terms of throughput and average delay. The results reveal that ACO based routing performs better than Dijkstra's routing.