Stochastic modeling of a power-managed system: Construction and optimization

Qinru Qiu, Qing Wu, Massoud Pedram

Research output: Chapter in Book/Entry/PoemChapter

44 Scopus citations


The goal of a dynamic power management policy is to reduce the power consumption of an electronic system by putting system components into different states, each representing certain performance and power consumption level. The policy determines the type and timing of these transitions based on the system history, workload and performance constraints. In this paper, we propose a new abstract model of a power-managed electronic system. We formulate the problem of system-level power management as a controlled optimization problem based on the theories of continuous-time Markov decision processes and stochastic networks. This problem is solved exactly and efficiently using a `policy iteration' approach. Our method is compared with existing heuristic approaches for different workload statistics. Experimental results show that power management method based on Markov decision process outperforms heuristic approaches in terms of power dissipation savings for a given level of system performance.

Original languageEnglish (US)
Title of host publicationProceedings of the International Symposium on Low Power Electronics and Design, Digest of Technical Papers
PublisherIEEE Computer Society
Number of pages6
StatePublished - 1999
Externally publishedYes
EventProceedings of the 1999 International Conference on Low Power Electronics and Design (ISLPED) - San Diego, CA, USA
Duration: Aug 16 1999Aug 17 1999


OtherProceedings of the 1999 International Conference on Low Power Electronics and Design (ISLPED)
CitySan Diego, CA, USA

ASJC Scopus subject areas

  • General Engineering


Dive into the research topics of 'Stochastic modeling of a power-managed system: Construction and optimization'. Together they form a unique fingerprint.

Cite this