Control Plane Scalability of Software Defined Networking

Software Defined Networks (SDN) present a novel network architecture designed to facilitate scalability and management. The SDN network comprises a control plane core responsible for governing and monitoring all SDN devices, which operate using the OpenFlow signaling protocol. However, the scalability of SDN networks poses a challenge. This research paper investigates the impact of network topology and size on the performance of the SDN control plane. The study employs performance metrics to compare three Java-based SDN controllers: Open Network Operating System (ONOS), OpenDaylight (ODL), and Floodlight controllers. The paper assesses metrics such as topology discovery time and control plane message overhead. The comparison involves fat-tree and spine and leaf topologies. All simulations are conducted within a centralized Out-of-band SDN environment. The results of the study demonstrate that ONOS controllers outperform ODL and Floodlight controllers in terms of discovery time, while the ODL controller exhibits the lowest overhead.