AI-Driven Scalable Framework for Energy-Efficient SDN Configurations
TL;DR
This work proposes a Machine Learning (ML)-based algorithm that uses clustering and classification techniques to find energy-efficient configurations in Software-Defined Networks, achieving up to 53% energy savings while reducing execution time by up to 8,000 times.
Abstract
The fast-paced development of 5G and 6G technologies, together with increasing environmental concerns, highlights the urgent need for energy-efficient computer networks. A major challenge lies in reducing energy usage by dynamically adapting the number of active network devices to the actual traffic demand while still fulfilling performance requirements. Traditional approaches, based on Integer Linear Programming (ILP) and heuristic algorithms, face significant limitations in scalability and computational efficiency, particularly for large-scale networks. To address these challenges, this work proposes a Machine Learning (ML)-based algorithm that uses clustering and classification techniques to find energy-efficient configurations in Software-Defined Networks. Simulations on realistic network topologies demonstrate that our solution delivers results close to those achieved by the reference algorithms, and near-optimal performance when ILP-based solutions are used as reference, achieving up to 53% energy savings while reducing execution time by up to 8,000 times. These results highlight the potential of ML-driven approaches to enable scalable and energy-efficient network management.