Experimental Categorization of SDN Controllers using Open-Source Benchmarking Tools

Abstract

Software Defined Networking SDN has increased network programability and administrative ease. SDN provides centralised and fine-grained network control; it has now become a trend in industry and is being used in different fields of networking. Controllers are crucial to the network’s stability and scalability in SDN, and their performance is critical. This research looks at three SDN controllers: RYU, POX and NOX. Quality of Service QoS measures such as Flow Setup Latency, Initial Ping Delay , Round Trip Time , Throughput, and TCP and UDP Band width are used to evaluate the controllers’ performance. Different sized networks are emulated using the Mininet SDN simulator by adjusting the number of switches in linear and tree topolo gies and hosts in a single topology. In compared to POX and NOX, my research demonstrates that RYU’s performance is quite consistent and exhibits little variation as the number of network devices and network traffic increases.Software Defined Networking SDN has increased network programability and administrative ease. SDN provides centralised and fine-grained network control; it has now become a trend in industry and is being used in different fields of networking. Controllers are crucial to the network’s stability and scalability in SDN, and their performance is critical. This research looks at three SDN controllers: RYU, POX and NOX. Quality of Service QoS measures such as Flow Setup Latency, Initial Ping Delay , Round Trip Time , Throughput, and TCP and UDP Band width are used to evaluate the controllers’ performance. Different sized networks are emulated using the Mininet SDN simulator by adjusting the number of switches in linear and tree topolo gies and hosts in a single topology. In compared to POX and NOX, my research demonstrates that RYU’s performance is quite consistent and exhibits little variation as the number of network devices and network traffic increases.