DocumentCode
2868027
Title
Dynamic voltage scheduling using adaptive filtering of workload traces
Author
Sinha, Amit ; Chandrakasan, Anantha P.
Author_Institution
Dept. of Electr. Eng. & Comput. Sci., MIT, Cambridge, MA, USA
fYear
2001
fDate
2001
Firstpage
221
Lastpage
226
Abstract
An adaptive approach for dynamic voltage scheduling on processors is presented based on workload prediction by filtering a trace history. The effects of update frequency and filtering strategy on the energy savings is analyzed. A performance hit metric is defined and techniques to minimize energy under a given performance requirement are outlined. Our results demonstrate that up to two orders of magnitude energy savings is possible with dynamic voltage scheduling depending on workload statistics
Keywords
circuit optimisation; computer power supplies; least mean squares methods; prediction theory; processor scheduling; statistical analysis; adaptive filtering; dynamic voltage scheduling; energy savings; magnitude energy savings; update frequency; workload prediction; workload statistics; workload traces; Adaptive filters; Adaptive scheduling; Delay; Dynamic scheduling; File servers; Filtering; Frequency; Processor scheduling; Statistics; Voltage control;
fLanguage
English
Publisher
ieee
Conference_Titel
VLSI Design, 2001. Fourteenth International Conference on
Conference_Location
Bangalore
ISSN
1063-9667
Print_ISBN
0-7695-0831-6
Type
conf
DOI
10.1109/ICVD.2001.902664
Filename
902664
Link To Document