Optimized Prediction Framework for Improving Cost Effectiveness of the Software Defined Network

Software defined network (SDN) technology has been adopted by organizations to increase network flexibility and reduce costs. However, network administrators face challenges in predicting network behavior and optimizing network configurations to reduce costs, leading to suboptimal network performance and increased costs. Hence, this paper proposes a novel Buffalo based You Only Look Once (Yolo) prediction framework (BbYPF) which was the combination of Buffalo Optimization algorithm and You Only Look Once (Yolo) prediction Framework. The required numbers of nodes were created in the SDN environment. Using the BbYPF, the required resources for each node in the SDN framework was estimated and allocated. Then, the optimal path is found by analyzing the trust node and nearest hub depends on the power consumption. It was implemented in the SDN framework for maximizing the cost effectiveness of the SDN system. The framework uses machine learning algorithms to analyze network data, make predictions about network behavior, and provide recommendations for network administrators on how to optimize network configurations to reduce costs. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.