Key points are not available for this paper at this time.
The main contribution of this paper is a novel hierarchi-cal scheme for adaptive dynamic power management (DPM) under nonstationary service requests. We model the non-stationary arrival process of service requests as a Markov-modulated stochastic process in which the stochastic process for eachmodulation statemodels a particular stationarymode of the arrival process. The bottom layer of our hierarchical ar-chitecture is a set of stationary optimal DPM policies, pre-calculated o-line for selected modes from policy optimization in Markov decision processes. The supervisory power man-ager at the top layer adaptively and optimally switches among these stationary policies on-line to accommodate the actual mode-switching arrival dynamics. Simulation results show that our approach, under highly nonstationary requests, can lead to signicant power savings compared to previously pro-posed heuristic approaches.
Ren et al. (Mon,) studied this question.