Designing a governor policy for energy saving and heat control in frequency-scaling green routers

Abstract

Energy costs for telecommunications networks are mainly due to the consumption of both devices and cooling facilities. For this reason the target of this paper is to propose an analytical discrete-time Markov model that allows green router designers to both evaluate performance of temperature-constrained green routers and design Governor policies to achieve the best trade-off between quality of service and energy saving in respect of a given target on the working temperature. The proposed model is applied to a case study to show how it can be used to the above purposes.