Dynamic power management for nonstationary service requests

Eui-Young Chung, Luca Benini, Alessandro Bogliolo, Yung Hsiang Lu, Giovanni De Micheli

Research output: Contribution to journalArticle

129 Citations (Scopus)

Abstract

Dynamic Power Management (DPM) is a design methodology aiming at reducing power consumption of electronic systems by performing selective shutdown of idle system resources. The effectiveness of a power management scheme depends critically on an accurate modeling of service requests and on the computation of the control policy. In this work, we present an online adaptive DPM scheme for systems that can be modeled as finite-state Markov chains. Online adaptation is required to deal with initially unknown or nonstationary workloads, which are very common in real-life systems. Our approach moves from exact policy optimization techniques in a known and stationary stochastic environment and it extends optimum stationary control policies to handle the unknown and nonstationary stochastic environment for practical applications. We introduce two workload learning techniques based on sliding windows and we study their properties. Furthermore, a two-dimensional interpolation technique is introduced to obtain adaptive policies from a precomputed look-up table of optimum stationary policies. The effectiveness of our approach is demonstrated by a complete DPM implementation on a laptop computer with a power-manageable hard disk that compares very favorably with existing DPM schemes.

Original languageEnglish
Pages (from-to)1345-1361
Number of pages17
JournalIEEE Transactions on Computers
Volume51
Issue number11
DOIs
Publication statusPublished - 2002 Nov 1

Fingerprint

Power Management
Control Policy
Workload
Unknown
Laptop computers
Adaptive Dynamics
Hard disk storage
Sliding Window
Look-up Table
Markov processes
Optimization Techniques
Power Consumption
Design Methodology
Markov chain
Interpolation
Electric power utilization
Interpolate
Power management
Electronics
Resources

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computational Theory and Mathematics

Cite this

Chung, Eui-Young ; Benini, Luca ; Bogliolo, Alessandro ; Lu, Yung Hsiang ; De Micheli, Giovanni. / Dynamic power management for nonstationary service requests. In: IEEE Transactions on Computers. 2002 ; Vol. 51, No. 11. pp. 1345-1361.
@article{68444eb9b84841459ce9433bf00338d0,
title = "Dynamic power management for nonstationary service requests",
abstract = "Dynamic Power Management (DPM) is a design methodology aiming at reducing power consumption of electronic systems by performing selective shutdown of idle system resources. The effectiveness of a power management scheme depends critically on an accurate modeling of service requests and on the computation of the control policy. In this work, we present an online adaptive DPM scheme for systems that can be modeled as finite-state Markov chains. Online adaptation is required to deal with initially unknown or nonstationary workloads, which are very common in real-life systems. Our approach moves from exact policy optimization techniques in a known and stationary stochastic environment and it extends optimum stationary control policies to handle the unknown and nonstationary stochastic environment for practical applications. We introduce two workload learning techniques based on sliding windows and we study their properties. Furthermore, a two-dimensional interpolation technique is introduced to obtain adaptive policies from a precomputed look-up table of optimum stationary policies. The effectiveness of our approach is demonstrated by a complete DPM implementation on a laptop computer with a power-manageable hard disk that compares very favorably with existing DPM schemes.",
author = "Eui-Young Chung and Luca Benini and Alessandro Bogliolo and Lu, {Yung Hsiang} and {De Micheli}, Giovanni",
year = "2002",
month = "11",
day = "1",
doi = "10.1109/TC.2002.1047758",
language = "English",
volume = "51",
pages = "1345--1361",
journal = "IEEE Transactions on Computers",
issn = "0018-9340",
publisher = "IEEE Computer Society",
number = "11",

}

Chung, E-Y, Benini, L, Bogliolo, A, Lu, YH & De Micheli, G 2002, 'Dynamic power management for nonstationary service requests', IEEE Transactions on Computers, vol. 51, no. 11, pp. 1345-1361. https://doi.org/10.1109/TC.2002.1047758

Dynamic power management for nonstationary service requests. / Chung, Eui-Young; Benini, Luca; Bogliolo, Alessandro; Lu, Yung Hsiang; De Micheli, Giovanni.

In: IEEE Transactions on Computers, Vol. 51, No. 11, 01.11.2002, p. 1345-1361.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Dynamic power management for nonstationary service requests

AU - Chung, Eui-Young

AU - Benini, Luca

AU - Bogliolo, Alessandro

AU - Lu, Yung Hsiang

AU - De Micheli, Giovanni

PY - 2002/11/1

Y1 - 2002/11/1

N2 - Dynamic Power Management (DPM) is a design methodology aiming at reducing power consumption of electronic systems by performing selective shutdown of idle system resources. The effectiveness of a power management scheme depends critically on an accurate modeling of service requests and on the computation of the control policy. In this work, we present an online adaptive DPM scheme for systems that can be modeled as finite-state Markov chains. Online adaptation is required to deal with initially unknown or nonstationary workloads, which are very common in real-life systems. Our approach moves from exact policy optimization techniques in a known and stationary stochastic environment and it extends optimum stationary control policies to handle the unknown and nonstationary stochastic environment for practical applications. We introduce two workload learning techniques based on sliding windows and we study their properties. Furthermore, a two-dimensional interpolation technique is introduced to obtain adaptive policies from a precomputed look-up table of optimum stationary policies. The effectiveness of our approach is demonstrated by a complete DPM implementation on a laptop computer with a power-manageable hard disk that compares very favorably with existing DPM schemes.

AB - Dynamic Power Management (DPM) is a design methodology aiming at reducing power consumption of electronic systems by performing selective shutdown of idle system resources. The effectiveness of a power management scheme depends critically on an accurate modeling of service requests and on the computation of the control policy. In this work, we present an online adaptive DPM scheme for systems that can be modeled as finite-state Markov chains. Online adaptation is required to deal with initially unknown or nonstationary workloads, which are very common in real-life systems. Our approach moves from exact policy optimization techniques in a known and stationary stochastic environment and it extends optimum stationary control policies to handle the unknown and nonstationary stochastic environment for practical applications. We introduce two workload learning techniques based on sliding windows and we study their properties. Furthermore, a two-dimensional interpolation technique is introduced to obtain adaptive policies from a precomputed look-up table of optimum stationary policies. The effectiveness of our approach is demonstrated by a complete DPM implementation on a laptop computer with a power-manageable hard disk that compares very favorably with existing DPM schemes.

UR - http://www.scopus.com/inward/record.url?scp=0036859651&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036859651&partnerID=8YFLogxK

U2 - 10.1109/TC.2002.1047758

DO - 10.1109/TC.2002.1047758

M3 - Article

AN - SCOPUS:0036859651

VL - 51

SP - 1345

EP - 1361

JO - IEEE Transactions on Computers

JF - IEEE Transactions on Computers

SN - 0018-9340

IS - 11

ER -