FANIC: FArthest Node Initialization Clustering technique for Controller Placement Problem in Software Defined Networking

Software Defined Networking is an emerging networking technology that simplifies the network management by separating the control logic from the underlying switches and routers. The control plane’s flexibility is heavily relied on the Controller Placement Problem (CPP), which deals with location of the controllers and assignment of forwarding elements to the controllers. Addressing the CPP should be the highest priority as this provides a significant improvement in reducing latency, addressing scalability and resilience, maintaining better load balancing and security in the network. Therefore, a new technique based on k-means clustering using Farthest Node Initialization (FANIC) has been suggested in order to assign the switches efficiently. This technique divides the network into k clusters and assigns all the switches to a specific controller. Then the best path from all the nodes to the controller is calculated inorder to reduce the latency between them. The suggested algorithm reduces the total controller-switch latency remarkably as compared to k-means clustering technique.