Presentation 2008-08-04
The Resolution of Trade-off between Power Consumption and Task Performance Using Elastic Task Model
Sayuri TERADA, Toshimitsu USHIO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In computing systems consisting of a CPU with dynamic voltage scaling (DVS), we can reduce their power consumption by setteing a low CPU frequency while performance of tasks are degraded. We generalize the elastic scheduling proposed by Marinoni and Buttazzo to resolve the trade-off between the power consumption and the task performance with desired rate. We evaluate the performance degradation by the increase of a force compressing elastic tasks where the elastic coefficient of each tasks indicates its importance. We propose an algorithm for minimizing an objective function given by weighted sum of the power consumption and the force, and its computational complexity is polynominal. We show some examples to demonstrate the efficiency of the algorithm.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) low power consumption / elastic task model / DVS
Paper # CST2008-15
Date of Issue

Conference Information
Committee CST
Conference Date 2008/7/28(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Concurrent System Technology (CST)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) The Resolution of Trade-off between Power Consumption and Task Performance Using Elastic Task Model
Sub Title (in English)
Keyword(1) low power consumption
Keyword(2) elastic task model
Keyword(3) DVS
1st Author's Name Sayuri TERADA
1st Author's Affiliation Graduate school of Engineering Science, Osaka University()
2nd Author's Name Toshimitsu USHIO
2nd Author's Affiliation Graduate school of Engineering Science, Osaka University
Date 2008-08-04
Paper # CST2008-15
Volume (vol) vol.108
Number (no) 176
Page pp.pp.-
#Pages 6
Date of Issue