Ryu-IDS: Intrusion Detection System for Modern Networks

As SDN continues to play a crucial role in modern network infrastructures, particularly in IoT-driven smart city applications, securing these environments against cyber threats is essential. This paper presents the deployment and evaluation of an Intrusion Detection System (IDS) within a Software-Defined Networking (SDN) environment, emphasizing its real-world impact on network performance. The IDS is implemented as an SDN-native application within the Ryu controller. A simulated SDN testbed is constructed using Mininet and Open vSwitch (OvS) to assess the IDS’s effect on key performance metrics such as latency, throughput, packet loss, CPU usage, and memory consumption. The results reveal that IDS increased network latency by an average of 0.016ms. Throughput decreased by approximately 100 kBps. Additionally, CPU usage rose by 5%, while memory usage increased by less than 1%. Unlike many existing studies that focus solely on Deep Learning (DL) model metrics, this research establishes a benchmark for evaluating the impact of DL-based IDS deployment in SDN environments.