Efficiency Assessment of Software-Defined Networking for Real-Time Network Systems

Abstract

: Software Networking (SDN) is growing in popularity due to its benefits, which include portability, mobility, analytics, and ease of creation. But it needs to be adequately shielded from security risks. One of the main vulnerabilities to the SDN network is the Distributed Denial-of-Service (DDoS) attack. To fulfill the demands of the complex and demanding security concerns of today, a new network philosophy is needed. Because SDN relies on a central controller, it has a single point of attack and failure. The present research monitors and analyses network traffic coming from switches, host computers, emulators, and wireless access points using a multi-vendor packet sampling technique using sFlow. In the process, we have clarified the usefulness and efficiency of the recommended strategy, which makes use of SDN controllers for the detection and mitigation of DDoS flooding attacks. The outcomes also demonstrate that the ODL controller outperforms the other remaining controllers in terms of load-shedding efficiency and flow setup latency. According to TCP bandwidth measurements, the ODL controller performs better in terms of processing power and jitter than the remaining controllers due to its higher computational complexity. These test results indicate that the jitter performance of all controllers is comparable. Overall analysis indicates that ODL is more reliable than other controllers in our scenario.