Securing Distributed SDN Controller Network from Induced DoS Attacks

摘要

With the escalation in security breaches on organizations and institutions that store, maintain and work with critical data, there is a need for a security-enhancing and risk-mitigating solution that works on the fly and is feasible to implement. Software Defined Networks is a networking paradigm that makes the network agile by disaggregating hardware and software. SDN helps enhance security with the help of micro-segmentation. The controller maintains a central view of the network, and its ability to monitor and store network information helps optimize routing. The centralization nature of the controller makes it vulnerable to DoS attacks which can be catastrophic for network functioning. The objective of this paper is to secure the distributed SDN controller architecture against DoS attacks. The proposed architecture is robust, scalable, and uses Big Data techniques to process streams of network traffic in real-time and Machine Learning to detect and mitigate DoS attacks.