J. Huang, T. Chen, Y. Lian, and H. Dai (PRC)
DPM, Flow Chart, Markov decision process, GSPN, Poisson process
In this paper, we present a new technique for modeling and solving the dynamic power management (DPM) problem for systems with complex behavioral characteristics such as concurrency, synchronization, mutual exclusion and conflict. We model a power-managed distributed computing system as a controllable Flow Chart. And then turn it to Generalized Stochastic Petri Net (GSPN) with cost. We use the Poisson process to optimization. The obtained GSPN model is automatically converted to an equivalent continuous-time Markov decision process. Given the delay constraints, the optimal power management policy for system components is calculated by solving a linear programming problem based on the Markov decision process, such as the optimal dispatch policy for requests. Experimental results show that the proposed technique can achieve more than 12% power saving compared to other existing DPM techniques.
Important Links:
Go Back
