SOA-MCP: A Nature Inspired Approach to Optimize Multiple Controller Placement in SDUWSNs

The seperation of the control plane from data plane has specific advantages in the context of Software Defined Networking (SDN). In software-defined underwater wireless sensor networks (SDUWSNs), the Controller Placement Problem (CPP) involves strategically placing controllers to meet essential criteria such as energy conservation, load distribution, latency and reduced processing times. This problem becomes more severe in extensive networks where scalability and efficient management are crucial. CPP is viewed as a multi-objective connective optimization task, which required fine balance among various optimization goals. This study proposes the innovative use of the Seagull Optimization Algorithm, a population-based meta-heuristic strategy, to enhance controller placement with a focus on minimizing Control Node - Controller and Controller - Controller latency and ensuring equitable load distribution across controllers. The efficacy of this strategy is evaluated by simulating on ns-3 with ONOS controller as the SDN policy provider. The simulation comparisons suggest that the proposed approach outperforms traditional methods and OpenFlow especially in the challenging conditions of underwater sensor networks.