Energy Aware Race to Halt: A Down to EARtH Approach for Platform Energy Management

Author

Efraim, Rotem ; Ginosar, Ran ; Weiser, C. ; Mendelson, Avi

Volume

13

Issue

1

fYear

2014

fDate

Jan.-June 18 2014

Firstpage

25

Lastpage

28

Abstract

The EARtH algorithm finds the optimal voltage and frequency operational point of the processor in order to achieve minimum energy of the computing platform. The algorithm is based on a theoretical model employing a small number of parameters, which are extracted from real systems using off-line and run-time methods. The model and algorithm have been validated on real systems using 45nm, 32nm and 22nm Intel® Core processors. The algorithm can save up to 44% energy compared with the commonly used fixed frequency policies.

Keywords

microprocessor chips; power aware computing; EARtH algorithm; Intel core processors; energy aware race to halt; fixed frequency policies; frequency operational point; off-line methods; optimal voltage; platform energy management; run-time methods; size 22 nm; size 32 nm; size 45 nm; Algorithm design and analysis; Computational modeling; Earth; Energy management; Energy measurement; Frequency measurement; Voltage measurement; B Hardware; B.9 Power Management; B.9.2 Energy-aware systems; C Computer Systems Organization; C.4 Performance of Systems; C.5 Computer System Implementation; C.5.4 VLSI Systems; C.5.5 Servers; Heterogeneous cores; Power Management;

fLanguage

English

Journal_Title

Computer Architecture Letters

Publisher

ieee

ISSN

1556-6056

Type

jour

DOI

10.1109/L-CA.2012.32

Filename

6327195