Artificial Intelligence Based Reliable Load Balancing Framework in Software-Defined Networks

Software-defined networking (SDN) plays a critical role in transforming networking from traditional to intelligent networking. The increasing demand for services from cloud users has increased the load on the network. An efficient system must handle various loads and increasing needs representing the relationships and dependence of businesses on automated measurement systems and guarantee the quality of service (QoS). The multiple paths from source to destination give a scope to select an optimal path by maintaining an equilibrium of load using some best algorithms. Moreover, the requests need to be transferred to reliable network elements. To address SDN’s current and future challenges, there is a need to know how artificial intelligence (AI) optimization techniques can efficiently balance the load. This study aims to explore two artificial intelligence optimization techniques, namely Ant Colony Optimization (ACO) and Particle Swarm Optimization (PSO), used for load balancing in SDN. Further, we identified that a modification to the existing optimization technique could improve the performance by using a reliable link and node to form the path to reach the target node and improve load balancing. Finally, we propose a conceptual framework for SDN futurology by evaluating node and link reliability, which can balance the load efficiently and improve QoS in SDN.