Load Balancing in Software Defined Networks

Abstract

Software Defined Network (SDN) has become a popular standard because of its advantages of centralized controlling and programmability for large-scale networks. In the world of digitization, high network traffic that needs to be handled optimally makes the visualization of whole network extremely complex. Over the years, the amount of traffic flowing through the network has increased exponentially and it is impossible for a single server to handle the huge incoming requests from the client nodes. Load balancing in SDN is used to manage the network traffic by partitioning the load among several servers in order to enhance the overall system efficiency. In this paper, we aim to develop a load balancing approach-using SDN that is essential in World Wide Web for providing high quality service. We thereby propose a Weighted Round Robin scheme that hands out the incoming requests among several servers using three network parameters like amount of data, transmit time and transfer time. The performance of the proposed approach is evaluated and compared with traditional Round Robin load balancing approach using mininet emulation. The results are evaluated on basis of response time (sec), transactions rate (trans/sec) and throughput (Mbit/sec). The results reveal better performance of the proposed scheme as compared to the reference approach.