DocumentCode :
2052498
Title :
An ISO-Energy-Efficient Approach to Scalable System Power-Performance Optimization
Author :
Song, Shuaiwen ; Grove, Matthew ; Cameron, Kirk W.
Author_Institution :
SCAPE Lab., Virginia Tech, Blacksburg, VA, USA
fYear :
2011
fDate :
26-30 Sept. 2011
Firstpage :
262
Lastpage :
271
Abstract :
The power consumption of a large scale system ultimately limits its performance. Consuming less energy while preserving performance leads to better system utilization at scale. The iso-energy-efficiency model was proposed as a metric and methodology for explaining power and performance efficiency on scalable systems. For use in practice, we need to determine what parameters should be modified to maintain a desired efficiency. Unfortunately, without extension, the iso-energy-efficiency model cannot be used for this purpose. In this paper we extend the iso-energy-efficiency model to identify appropriate efficiency values for workload and power scaling on clusters. We propose the use of "correlation functions" to quantitatively explain the isolated and interacting effects of these two parameters for three representative applications: LINPACK, row-oriented matrix multiplication, and 3D Fourier transform. We show quantitatively that the iso-energy-efficiency model with correlation functions is effective at maintaining efficiency as system size scales.
Keywords :
Fourier transforms; matrix algebra; optimisation; performance evaluation; power aware computing; 3D Fourier transform; LINPACK; correlation functions; iso-energy efficient approach; power consumption; power scaling; row oriented matrix multiplication; scalable system power performance optimization; Analytical models; Computational modeling; Computer architecture; Correlation; Energy consumption; Integrated circuit modeling; System-on-a-chip; Iso-energy-efficiency; performance isoefficiency; power aware computing; system utilization;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Cluster Computing (CLUSTER), 2011 IEEE International Conference on
Conference_Location :
Austin, TX
Print_ISBN :
978-1-4577-1355-2
Electronic_ISBN :
978-0-7695-4516-5
Type :
conf
DOI :
10.1109/CLUSTER.2011.37
Filename :
6061144
Link To Document :
بازگشت