Energy Optimization for Scientific Programs Using Auto-tuning Language ppOpen-AT

Author

Katagiri, Takahiro ; Cheng Luo ; Suda, Ryutaro ; Hirasawa, Shoichi ; Ohshima, Shigetoshi

Author_Institution

Inf. Technol. Center, Univ. of Tokyo, Tokyo, Japan

fYear

2013

fDate

26-28 Sept. 2013

Firstpage

123

Lastpage

128

Abstract

In this paper, we demonstrate a new approach for power-consumption optimization using a dedicated Auto-tuning (AT) language. Our approach is based on recently developed technologies: (1) a power measurement application programming interface, (2) an AT mathematical core library. Preliminary performance evaluation enables us to select the best kernel for a real-world scientific program using either the CPU or Graphics Processing Unit, with respect to energy consumption. From the results of the evaluation, we found the performance-changing point in the experimental environment.

Keywords

application program interfaces; energy conservation; graphics processing units; natural sciences computing; performance evaluation; power aware computing; program compilers; software libraries; AT mathematical core library; CPU; dedicated autotuning language; energy consumption; energy optimization; graphics processing unit; kernel selection; performance evaluation; performance-changing point; power consumption optimization; power measurement application programming interface;; ppOpen-AT; real-world scientific program; standard API; Educational institutions; Fluctuations; Graphics processing units; Kernel; Optimization; Power demand; Power measurement; ATMathCoreLib; Auto-tuning; Optimization; Standard API; ppOpen-AT;

fLanguage

English

Publisher

ieee

Conference_Titel

Embedded Multicore Socs (MCSoC), 2013 IEEE 7th International Symposium on

Conference_Location

Tokyo

Type

conf

DOI

10.1109/MCSoC.2013.14

Filename

6657916