A Comparative Analysis of SDN Controller Placement Problem: IoT-Specific Tactics and Generalized Solutions
Tangudu Harsha Vardhan, Yellina SriBhargav, Jahnavi Anala, K Yejnakshari Meghana, N. Radhika · 2025
Software-Defined Networking (SDN) revolutionizes network management by decoupling control and data planes, presenting the challenge of the Controller Placement Problem (CPP). This paper compares various approaches to CPP, focusing on strategies for IoT environments and general SDN networks. The IoT-centric study highlights the importance of adaptive, energy-efficient placements using heuristic methods like Particle Swarm Optimization, tailored to resourceconstrained scenarios. In contrast, broader SDN strategies involve computationally intensive methods like Integer Linear Programming and machine learning for large-scale networks. Identified research gaps include the need for dynamic realtime placement, energy-efficient solutions, and multi-domain coordination. The paper proposes enhancements using reinforcement learning and collaborative approaches to optimize controller placement. These solutions aim to improve network performance, scalability, and fault tolerance in diverse SDN applications, providing a foundation for future research in SDN architecture optimization.