Optimized Controller Placement and Load Balancing in SDN-Enabled IoT Networks

doi:10.1007/s00542-025-06004-x

Optimized Controller Placement and Load Balancing in SDN-Enabled IoT Networks

2026-03-25

1

PRE

AI

DOI：10.1007/s00542-025-06004-x

原文链接

原文求助

摘要

En 中文

The convergence of Software-Defined Networking (SDN) and the Internet of Things (IoT) has enabled flexible and scalable network management but has also introduced complex challenges such as optimal controller placement and dynamic load balancing. This paper presents an Optimized Controller Placement and Load Balancing (OCP-LB) framework tailored for SDN-enabled IoT networks. The proposed solution adopts a two-stage strategy: (1) identification of critical controller locations using a k-core decomposition-based articulation point method, and (2) optimization of controller placement using the Artificial Protozoa Optimization Algorithm (APOA). Furthermore, a dynamic load balancing and traffic flow management algorithm ensures efficient controller-to-switch mapping under varying traffic conditions and failure scenarios. Simulation results demonstrate that OCP-LB significantly outperforms existing metaheuristic approaches such as Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO). Specifically, latency is reduced by up to 21%, fault tolerance improves by 14.3%, load variance is minimized by 28.9%, and average jitter is lowered by 45.5%. Additionally, OCP-LB achieves up to 27.4% higher throughput, 36.4% better energy efficiency, and a 35% faster convergence speed. These results establish OCP-LB as a robust and scalable solution for enhancing reliability, performance, and adaptability in large-scale IoT-SDN environments.

Optimized Controller Placement and Load Balancing in SDN-Enabled IoT Networks

摘要

期刊

机构

引用论文