Journal of Engineering Research and Reports

This research examines the vulnerabilities and resilience mechanisms of Software-Defined Networking (SDN) and cloud networks, with a specific focus on controller failures and security attacks. The study leverages both simulated and real-world data to assess how these vulnerabilities impact network performance metrics including downtime, packet loss, latency, and throughput. A significant observation from the study is that the nature and impact of network disruptions vary significantly depending on the type of failure or attack, highlighting the need for tailored resilience strategies. Machine learning techniques, notably Support Vector Machines (SVMs), are employed to classify these disruptions with high accuracy, suggesting a promising direction for proactive network management. The research proposes a novel framework that combines the dynamic control capabilities of SDN with machine learning and automation to improve the networks’ fault tolerance and recovery mechanisms. The effectiveness of this framework is demonstrated through enhanced resilience and reduced performance degradation during network disruptions. This study contributes to the field by outlining a scalable and efficient approach to mitigating vulnerabilities in SDN and cloud networks, thereby enhancing overall network stability and reliability.