M.A. Thesis:

Proposing an Algorithm with Load Balancing on Software Defined Networks based On Grey Wolf Optimization Algorithm

Software-Defined networks are a novel network architecture in which the data plane is separated from the control plane by the OpenFlow protocol. The main characteristic of these networks is the centralized control of the network for resource allocation. As a result, all of the key decisions about packet forwarding are made in the network controller, and data-level tools only drive packets based on those decisions. Link-load balancing is a model for identifying the best path between multiple nodes in a network; this identification is done to maximize resource utilization and minimize costs (distance, transmission time, etc.). Also, load balancing plays an effective role in preventing load loss and providing QoS assurance. On the other hand, load balancing can provide maximum network productivity and resource balance [3].The Gray Wolf Optimization algorithm is a meta-heuristic, and nature-based algorithm that is regulated by the hierarchical structure and social behavior of wolves during hunting. This algorithm is population-based and has been used in recent years to solve many optimization problems. Some studies have shown that this algorithm yields better results than some other algorithms [7]. Load balancing algorithms are inherently part of the routing problem and, in terms of hardness, are part of the NP-Complete problem. In recent years, meta-heuristic algorithms have been proposed to solve this type of problem and to find an optimal solution in acceptable time. In this research, we are using one of these meta-heuristic algorithms (GWO) to provide an optimal solution for load balancing problem in Software-Defined networks.

Keywords: