Description

Software Defined Networks (SDN) is a new ideology which has recently engrossed network researchers working under programmable networks by reassuring to make the process of management and designing of networks more simplified and innovative compared to the already existing and established but intransigent current approach. The network partitioning or dividing with SDN control planes addresses issues such as reliability, latency and load balancing. This results in the introduction of the problem of controller placement, which has become a major area of concern in SDNs. We consider both the controller to router and controller to controller latency and attempt to provide solutions to minimize the total latency. We also consider maximizing the fault tolerance. We adopt stochastic meta-heuristic algorithms called the Particle Swarm Optimization and Firefly Algorithm for the same. The analysis should help shed some light on the effectiveness of these algorithms for solving discrete optimization problems and should be useful to professionals working in networking and optimization domain or anyone else with an interest in the amalgamation of networking and meta-heuristic optimization.