Performance analysis of software defined network controller architecture—A simulation based survey

Abstract

Software Defined Networking (SDN) has proclaimed significant attention from network researcherers and industries in recent years. Software defined network is an effort to design, build, and manage networks that separates the networks control (brains) and data (muscle) planes. It enables the network control to become directly programmable and the underlying architecture to be abstracted for applications and network services. SDN platform provides various advantages such as programmability, potential for task virtualization and schedulability and easy management of the network. However, it introduces new challenges towards scalability and performances, security hardening, cross-layer communication protocols, etc. It is important to understand and analyze the performances and limitations of SDN for implementation and deployment in live network environments and applications. This paper reports a number of technologies, models and tools to evaluate the performance metrics of SDN controllers along with simulation results. It also states the working procedure of various controllers like NOX, NOX-MT, POX, Rosemary, FloodLight, OpenDayLight, Beacon, Maestro. This study will help in designing efficient architecture and control algorithms for SDN controllers.