Mohammad Kazemiesfeh ,Somaye Imanpour,Ahmadreza Montazerolghaem
Date published: 1 April 2025
Published by Elsevier B.V.

Abstract

Deploying multiple controllers in the control panel of software-defined networks increases scalability, availability, and performance, but it also brings challenges, such as controller overload. To address this, load-balancing techniques are employed in software-defined networks.Controller load balancing can be categorized into two main approaches: (1) single-level thresholds and (2) multi-level thresholds. However, previous studies have predominantly relied on single-level thresholds, which result in an imprecise classification of controllers or have assumed uniform controller capacities in multi-level threshold methods. This study explores controller load balancing with a focus on utilizing multi-level thresholds to accurately assess controller status. Switch migration operations are utilized to achieve load balancing, considering factors such as the degree of load imbalance of the target controller and migration efficiency. This includes evaluating the post-migration status of the target controller and the distance between the migrating switch and the target controller to select the appropriate target controller and migrating switch. The proposed scheme reduces controller response time, migration costs, communication overhead, and throughput rate. Results demonstrate that our scheme outperforms others regarding response time and overall performance.