Double Level Code Scanning Leveraging Network virtualization and Machine Learning to Mitigate Memory Starvation Risks and Insecure Programming

Abstract

In this study, we investigate the role of software-defined networking (SDN) controller on memory usage, system stability, and performance in cloud computing networks, with a specific focus on the deployment of machine learning techniques. The SDN controller utilizes machine learning algorithms to dynamically detect and block vulnerable and memory-intensive cloudlet codes. The results demonstrate that the deployment of machine learning in the SDN controller significantly improves resource allocation, mitigates resource contention, and reduces the risk of memory exhaustion. By accurately identifying and preventing the execution of insecure cloudlets, the machine learning-enabled SDN controller enhances system security and maintains optimal resource utilization. Moreover, the study highlights the positive influence of the SDN controller on system stability, preventing disruptive operations and minimizing the likelihood of system crashes or slowdowns.