Aggressive look-ahead earliest deadline first algorithm

Lih-Yih Chiou; Hsin Ei Lim; Yi Siou Chen

doi:10.1109/TENCON.2007.4428980

Aggressive look-ahead earliest deadline first algorithm

Lih-Yih Chiou, Hsin Ei Lim, Yi Siou Chen

Department of Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Dynamic voltage and frequency scaling technology is a promising way to save system energy. We proposed an energy-efficient scaling algorithm which utilizes both static and dynamic slack to adjust the frequency and voltage of a processor for running periodic tasks. Instead of excessively using up slack for the task under consideration, our proposed algorithm takes a global approach by intelligently assigning proper slices of slack to tasks and scaling the processor correspondingly. Experiments with an example system demonstrate that the energy consumption of our proposed algorithm is not larger than 3% of the theoretical limit.

Original language	English
Title of host publication	TENCON 2007 - 2007 IEEE Region 10 Conference
DOIs	https://doi.org/10.1109/TENCON.2007.4428980
Publication status	Published - 2007 Dec 1
Event	IEEE Region 10 Conference, TENCON 2007 - Taipei, Taiwan Duration: 2007 Oct 30 → 2007 Nov 2

Publication series

Name	IEEE Region 10 Annual International Conference, Proceedings/TENCON

Other

Other	IEEE Region 10 Conference, TENCON 2007
Country	Taiwan
City	Taipei
Period	07-10-30 → 07-11-02

All Science Journal Classification (ASJC) codes

Computer Science Applications
Electrical and Electronic Engineering

Access to Document

10.1109/TENCON.2007.4428980

Fingerprint Dive into the research topics of 'Aggressive look-ahead earliest deadline first algorithm'. Together they form a unique fingerprint.

Cite this

@inproceedings{99a30b03c9314b4ca79c991b9fcd4b42,

title = "Aggressive look-ahead earliest deadline first algorithm",

abstract = "Dynamic voltage and frequency scaling technology is a promising way to save system energy. We proposed an energy-efficient scaling algorithm which utilizes both static and dynamic slack to adjust the frequency and voltage of a processor for running periodic tasks. Instead of excessively using up slack for the task under consideration, our proposed algorithm takes a global approach by intelligently assigning proper slices of slack to tasks and scaling the processor correspondingly. Experiments with an example system demonstrate that the energy consumption of our proposed algorithm is not larger than 3% of the theoretical limit.",

author = "Lih-Yih Chiou and Lim, {Hsin Ei} and Chen, {Yi Siou}",

year = "2007",

month = dec,

day = "1",

doi = "10.1109/TENCON.2007.4428980",

language = "English",

isbn = "1424412722",

series = "IEEE Region 10 Annual International Conference, Proceedings/TENCON",

booktitle = "TENCON 2007 - 2007 IEEE Region 10 Conference",

note = "IEEE Region 10 Conference, TENCON 2007 ; Conference date: 30-10-2007 Through 02-11-2007",

}

TY - GEN

T1 - Aggressive look-ahead earliest deadline first algorithm

AU - Chiou, Lih-Yih

AU - Lim, Hsin Ei

AU - Chen, Yi Siou

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Dynamic voltage and frequency scaling technology is a promising way to save system energy. We proposed an energy-efficient scaling algorithm which utilizes both static and dynamic slack to adjust the frequency and voltage of a processor for running periodic tasks. Instead of excessively using up slack for the task under consideration, our proposed algorithm takes a global approach by intelligently assigning proper slices of slack to tasks and scaling the processor correspondingly. Experiments with an example system demonstrate that the energy consumption of our proposed algorithm is not larger than 3% of the theoretical limit.

AB - Dynamic voltage and frequency scaling technology is a promising way to save system energy. We proposed an energy-efficient scaling algorithm which utilizes both static and dynamic slack to adjust the frequency and voltage of a processor for running periodic tasks. Instead of excessively using up slack for the task under consideration, our proposed algorithm takes a global approach by intelligently assigning proper slices of slack to tasks and scaling the processor correspondingly. Experiments with an example system demonstrate that the energy consumption of our proposed algorithm is not larger than 3% of the theoretical limit.

UR - http://www.scopus.com/inward/record.url?scp=48649089793&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=48649089793&partnerID=8YFLogxK

U2 - 10.1109/TENCON.2007.4428980

DO - 10.1109/TENCON.2007.4428980

M3 - Conference contribution

AN - SCOPUS:48649089793

SN - 1424412722

SN - 9781424412723

T3 - IEEE Region 10 Annual International Conference, Proceedings/TENCON

BT - TENCON 2007 - 2007 IEEE Region 10 Conference

T2 - IEEE Region 10 Conference, TENCON 2007

Y2 - 30 October 2007 through 2 November 2007

ER -