A Secure Switch Migration Scheduling based on Prediction for Load Balancing in SDN

Abstract

The control plane of Software Defined Networking (SDN) Featured in the distributed architecture to achieve the scalability, thus the load is divided between the controllers. each controller must have the appropriate load. so, the control plane needs an accurate and beneficial design to meet the load balancing between the controllers. Thus, the switch migration is the most proposed solution for load balancing in SDN. Most of the proposed algorithms are implemented to produce efficient switch migration. In this paper, we propose a secure switch migration scheduling algorithm to enhance the load balancing and achieve the security between the switch and the destination controller. Our algorithm uses an ARIMA prediction approach to predict the load of the controllers and then according to the prediction, the switch migration operation is performed. It reduces the load for the controller which has a very large load by offloading it quickly and it makes the open-flow switch distinguishes if the destination controller is malicious or not. The results show that our algorithm enhances the load balancing compared to [10] and it confirms the security between the switch and the destination controller.