Graph Detection Time of Software Defined Network

Software Defined Networks (SDNs) propose the decoupling of the control plane from the forwarding plane. To implement this new architecture, the OpenFlow SDN control protocol can be employed for communication between virtual switches. This approach enables centralized management and monitoring. However, the development of SDN technology necessitates appropriate testing and measurement tools to assess performance. Currently available open-source performance tools for SDN architecture primarily focus on basic attributes, overlooking the evaluation of numerous SDN controllers in terms of Graph Detection time. This study aims to address this gap by introducing a performance monitoring tool for SDN controllers, which evaluates two Javabased SDN controllers. The tool measures the Graph Detection speed across various network sizes using a Python-based northbound interface. The evaluation includes popular controllers like Open Network Operating System and OpenDaylight, assessing their performance with dedicated control channels. Simulation results indicate that OpenDaylight outperforms Open Network Operating System in terms of Graph Detection time. Index Terms—Graph Detection Time, ODL Controller, ONOS Controller, Rest API, Dedicated Control Channel, Control Northbound Interface.