A network energy efficiency and load balance model based on nonlinear programming

In network research community, it has been considerably accepted that there is a trade-off between energy efficiency and load balance. However, there were very few solutions proposed to improve both of them equally. In this paper, we derive a nonlinear programming (NLP) model to settle the both aspects, by designing two nonlinear functions and introducing a nonlinear power-bandwidth function. The simulation results show that, our model achieves near-optimal in energy efficiency and load balance simultaneously. Comparing to open shortest path first (OSPF), our model reduces the maximum link utilization (MLU) by 54.7%, and the power consumption by 68.5%.