On the Performance of SDN Controllers in Real World Topologies

Software Defined Networking (SDN) represents a new paradigm in networking that shifts the traditional architecture from a fully distributed model to a centralized one. The central principle is the separation of the data and control layers. The logic and intelligence of the control layer is moved to the controller, a software tool, usually open source, used to develop, standardize, and manage network applications. It bridges the communication of network applications and the underlying infrastructure, providing programmable interfaces and other functionalities. In this work, the controllers Ryu, ONOS, OpenDalight and Floodlight were evaluated using the Mininet emulator and the iPerf tool. We focus on the Fat-tree topology, which is a complex topology widely used in data centers. The metrics taken are throughput, latency, and Jitter. The aim is to investigate and compare the performance of the four controllers, providing in-depth analysis and giving insights about their behavior and effectiveness.