Using bio-inspired algorithms for energy levels assessment in energy efficient wired communication networks

Rapid growth of ICT (Information Communication Technologies) energy consumption involves the need for proposing new mechanisms to enhance their energy efficiency. Focusing on energy consumption of networking equipment, this paper presents a study to achieve a tradeoff between the amount of energy that could be saved in wired networks and the discrete number of energy levels to be implemented by line cards. We use bio-inspired computing based on GA (Genetic Algorithms) and PSO (Particle Swarm Optimization) in order to assess the most suitable network configurations in terms of energy savings for different-sized networks such as NSFNet, Géant and AT&T. Results show a comparison between both bio-inspired algorithms in which, although GA produces better results, PSO achieves a reduction in computation time with an optimality gap below 1.7%. From a practical point of view, a limited number, such as four energy levels, is enough to achieve significant reductions in energy consumption.