Performance analysis of ODL and RYU controllers’ against DDoS attack in software defined networks

Performance analysis of ODL and RYU controllers’ against DDoS attack in software defined networks

연구 분야: Networking

학회: Cluster Computing

Software-defined networking (SDN) is a new way of designing and managing networks. The central SDN controller serves as the network’s brain in the control plane. This paper compares the performance of OpenDayLight (ODL) and Ryu SDN controllers by installing the most stable version of the controllers on a realistic test environment, Mininet, on which several other SDN controllers are built. The sFlow and snort tools have been used to compare the performance of the two controllers for a variety of different workloads. Tests were carried out by launching Distributed Denial of Service (DDoS) attacks on the host, stopping them, and then examining the outcomes. Experimental results show that the ODL controller is more effective than the Ryu controller in dealing with DDoS attacks on the SDN network. DDoS attacks are detected using Mininet analysis and the tools Snort and sFlow. The results also show that both controllers perform better in terms of flow setup latency and load shedding performance compared to the Ryu controller. The ODL controller exceeds the Ryu controller in levels of jitter, and its computational complexity outperforms the Ryu controller in terms of processing power. Based on these test findings, it can be concluded that both controllers perform equally well in terms of jitter and ODL is more reliable than Ryu.