RLA-ENAR: A realistic near-optimal energy-aware routing

Energy consumption is an important issue due to its environmental and economic effects; hence, recently there is significant effort to reduce energy consumption in all areas including computer networks, which is realized by the Green Networking concept. As an approach of green networking, some energy-aware routing algorithms have already been proposed in the literature. However, most on them are based on artificial power models and cannot reach the optimal performance because of the NP-Hardness of the problem. In this paper, by analyzing a realistic power model, we propose a two-phase routing algorithm to minimize network power consumption while satisfying a set of given traffic demands. The second contribution is the near-optimal performance of the algorithm. Simulation results in a real network topology and practical power consumption parameter settings, show that the algorithm achieves 40% energy saving in comparison with OSPF-TE and 18% in comparison with OSPF-EC. Comparing the result with the optimal solution which is obtained by an Integer Linear Programming model shows that the proposed algorithm gives near optimal solution.