Statistical power modeling of GPU kernels using performance counters

@article{Nagasaka2010StatisticalPM,
  title={Statistical power modeling of GPU kernels using performance counters},
  author={Hitoshi Nagasaka and Naoya Maruyama and Akira Nukada and Toshio Endo and Satoshi Matsuoka},
  journal={International Conference on Green Computing},
  year={2010},
  pages={115-122}
}
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… CONTINUE READING
Highly Cited
This paper has 123 citations. REVIEW CITATIONS

7 Figures & Tables

Topics

Statistics

0102030201020112012201320142015201620172018
Citations per Year

124 Citations

Semantic Scholar estimates that this publication has 124 citations based on the available data.

See our FAQ for additional information.