A Systematic Study and Detailed Performance Assessment of SDN Controllers Across a Wide Range of Network Architectures

Internet access is the primary motivating factor behind the revolutionary modern digital era. Almost everything is connected to the Internet because of the Internet of Things (IoT) concept. However, because typical IP networks have a close relationship between the data plane and the control plane, managing and configuring the network are extremely time-consuming and expensive to compute. Software-defined networking (SDN) has been recommended as a conceptual shift toward a perceived and functionally unified network control plane that will decrease the complexity of network administration. SDN makes it possible to program the network’s configuration and divides the data plane from the control plane. In the control plane, the controller supervises all data plane operations as its primary component. Therefore, for optimal performance, the controller’s performance and capabilities are crucial. The literature contains several controller designs for both broad and narrow networks have been proposed. Yet, there is just a little amount of thorough quantitative examination of them. In this article, a comprehensive qualitative comparison of various SDN controllers and a quantitative analysis of their performance in a variety of network scenarios are presented. We classify and categorize controllers, then compare them qualitatively. Several types of network controllers are compared and contrasted, including those used in IoT, blockchain, vehicle, and wireless sensor networks.