The Community for Technology Leaders
International Conference on Green Computing (2010)
Chicago, IL, USA
Aug. 15, 2010 to Aug. 18, 2010
ISBN: 978-1-4244-7612-1
pp: 115-122
Hitoshi Nagasaka , Tokyo Institute of Technology, Japan
Naoya Maruyama , Tokyo Institute of Technology, Japan
Akira Nukada , Tokyo Institute of Technology, Japan
Toshio Endo , Tokyo Institute of Technology, Japan
Satoshi Matsuoka , Tokyo Institute of Technology, Japan
We present a statistical approach for estimating power consumption of GPU kernels. We use the GPU performance counters that are exposed for CUDA applications, and train a linear regression model where performance counters are used as independent variables and power consumption is the dependent variable. For model training and evaluation, we use publicly available CUDA applications, consisting of 49 kernels in the CUDA SDK and the Rodinia benchmark suite. Our regression model achieves highly accurate estimates for many of the tested kernels, where the average error ratio is 4.7%. However, we also find that it fails to yield accurate estimates for kernels with texture reads because of the lack of performance counters for monitoring texture accesses, resulting in significant underestimation for such kernels.

A. Nukada, N. Maruyama, T. Endo, S. Matsuoka and H. Nagasaka, "Statistical power modeling of GPU kernels using performance counters," International Conference on Green Computing(GREENCOMP), Chicago, IL, USA, 2010, pp. 115-122.
82 ms
(Ver 3.3 (11022016))