Multi-Controllers Placement Optimization in SDN by the Hybrid HSA-PSO Algorithm

Abstract

Software-Defined Networking (SDN) is a developing architecture that provides scalability, flexibility, and efficient network management. However, optimal controller placement faces many problems, which affect the performance of the overall network. To resolve the Multi-controller SDN (MC-SDN) that is deployed in the SDN environment, we propose an approach that uses a hybrid metaheuristic algorithm that improves network performance. Initially, the proposed SDN network is constructed based on graph theory, which improves the connectivity and flexibility between switches and controllers. After that, the controller selection is performed by selecting an optimal controller from multiple controllers based on controller features using the firefly optimization algorithm (FA), which improves the network performance. Finally, multi-controller placement is performed to reduce the communication latency between the switch to controllers. Here, multiple controllers are placed by considering location and distance using a hybrid metaheuristic algorithm, which includes a harmonic search algorithm and particle swarm optimization algorithm (HSA-PSO), in which the PSO algorithm is proposed to automatically update the harmonic search parameters. The simulation of multi-controller placement is carried out by the CloudsimSDN network simulator, and the simulation results demonstrate the proposed advantages in terms of propagation latency, Round Trip Time (RTT), matrix of Time Session (TS), delay, reliability, and throughput.

Multi controller Placement optimization in Software Defined Network Controllers by using Genetic Algorithm

“…Another set of simulations have been done by performed for the evaluation of the proposed algorithm. The comparison with other existing techniques such as MC-SDN [40] Annealing Failure Foresight Capacitated Controller Placement Problem (SA-FFCCPP) [41] along with Garter Snake Optimization Capacitated Controller Placement Problem (GSOCCPP) [42], have made in order to evaluate the performance of proposed Ga based technique. All these above-mentioned techniques are simulated on the basis of many parameters such as propagation delay, average round-trip time (RTT), average delay, reliability, and throughput, respectively.…”

SDN-Based VANET Routing: A Comprehensive Survey on Architectures, Protocols, Analysis, and Future Challenges

“…Machine learning algorithms such as reinforcement learning (RL) can be employed to determine the optimal localization of SDN controllers, including the number and distribution of controllers [183]. However, none of the SDVN routing proposals address this issue which is still doubtful and needs more research effort.…”

Controller placement problem during SDN deployment in the ISP/Telco networks: A survey

“…Most popular open source SDN controllers in industry and academia includes the Open Network Operating System (ONOS), OpenDayLight (ODL), OpenKilda, Ryu and Faucet. 22 The modularity of each controller is governed by the design focus and programming languages. The controller's portability and performance are affected by the programming language choice.…”

“…The optimal selection of controllers is based on the controller features. Most popular open source SDN controllers in industry and academia includes the Open Network Operating System (ONOS), OpenDayLight (ODL), OpenKilda, Ryu and Faucet 22 …”