No abstract

The power consumption of core networks is bound to grow considerably due to increasing traffic volumes. Network reconfiguration adapting resources to the load is a promising countermeasure. However, the benefit of this approach is hard to evaluate realistically since current network equipment does not support dynamic resource adaptation and power-saving features. In this paper, we derive a dynamic resource operation and power model for IP-over-WSON network devices based on static power consumption data from vendors and reasonable assumptions on the achievable scaling behavior. Our model allows to express the dynamic energy consumption as a function of active optical interfaces, line cards, chassis, and the amount of switched IP traffic. We finally apply the model in the evaluation of two network reconfiguration schemes.

Due to increasing traffic volumes and access bandwidths, the power consumption of core networks will grow considerably. Adapting network configuration to traffic load is one counter-measure. Technological and operational limitations thereby require transitioning between subsequent configurations in a single step. In this paper, we define the according reconfiguration problem with resource-preoccupation constraints, and we propose and evaluate a meta-heuristics based solution method for this problem. We obtain energy savings of up to 35 % compared to a simple resource adaptation scheme and only observe rare events of traffic blocking in particular situations.