Load balancing in SDN based networks using additive increase and multiplicative decrease (AIMD) technique

The current trends in Internet based applications shows that the amount of data traffic flowing between networks have increased dramatically over the years. Recently emerged and growing applications such as - Cloud Computing, Big data and Internet of Things have increased the need for more powerful and fast networks that are capable of handling vast and diverse data. The traditional networks are not fast enough and efficient for handling such internet data traffic. Though Software defined Networking has proved to be efficient enough to handle the network traffic caused due to these demanding applications, the load each server can handle is limited. To overcome this problem the network service provider's use multiple servers with similar capabilities and functions. By balancing the load intelligently between these server replicas, the efficiency of the server is scaled up. This paper will describe a load balancing technique that uses various features of SDN and OpenFlow to create a load balancer application for Software Defined Networks.