Investigating the Impact of Latency on Network Performance in Software Defined Network: Simulation Based Approach

Software Defined Network is intelligent network in which the state of network is handled by controllers, and the underlying network infrastructure is abstracted from the applications. The communication between the controller and the OpenFlow switches uses the OpenFlow protocol which is responsible for forwarding packets as per the controller's instruction. The controller has a global view of the entire network which has a negative impact on the performance of the network. The end to end delay degrades the performance of the controllers. The decoupling of the control plane from the data plane causes latency and delays between the data plane and the control plane. Latency is the main challenge in the data centre and it causes delays in packet delivery and congestion. There is a need to further investigate the impact of latency on the performance of the network. In this study, we evaluated the performance of the OpenFlow controllers such as the NOX, POX, and Floodlight in MININET simulation. The schemes were evaluated based on delay, throughput, packet loss, and respond time of the controllers. The results show that the floodlight is the best performing controller.