A Modified Algorithm for Modelling SDN Data traffic

Abstract

Software Defined Networking (SDN) can allegedly maintain the quality of service (QoS) standards even in the presence of hostile flow, according to recent claims. Queueing theory, and more specifically network models, have been used for a very long time to examine the performance and QoS characteristics of networks in order to investigate this phenomenon. Because of the dependencies between the layers, planes, and components in an SDN architecture, the latter model seems especially well suited to represent the behavior of SDN. Numerous papers have described network models to examine the behavior of various network design applications. Here, we demonstrate how to employ the Markov-modulated Poisson process (MMPP) model to mathematically depict SDN traffic. Many articles had recommended utilizing MMPP to assess and model different types of IP network traffic, and it was widely used to simulate the traffic on traditional IP networks. We assert that MMPP can represent SDN traffic just as if it would in traditional IP networks. Our tests in this study indicate that MMPP is a useful technique for studying SDN data traffic. Starting with the premise that SDN traffic is averaged across multiple experiments and using two different SDN network topologies, we proceed. Emulation tests revealed that MMPP is a good model for SDN data traffic.