6904

Optimization of power consumption in the iterative solution of sparse linear systems on graphics processors

Hartwig Anzt, Maribel Castillo, Juan C. Fernandez, Vincent Heuveline, Francisco D. Igual, Rafael Mayo and Enrique S. Quintana-Orti
Institute for Applied and Numerical Mathematics 4, Karlsruhe Institute of Technology, Fritz-Erler-Str. 23, 76133 Karlsruhe, Germany
Computer Science – Research and Development, 2011

@article{springerlink:10.1007/s00450-011-0195-8,

   author={Anzt, Hartwig and Castillo, Maribel and Fernandez, Juan and Heuveline, Vincent and Igual, Francisco and Mayo, Rafael and Quintana-Orti, Enrique},

   affiliation={Institute for Applied and Numerical Mathematics 4, Karlsruhe Institute of Technology, Fritz-Erler-Str. 23, 76133 Karlsruhe, Germany},

   title={Optimization of power consumption in the iterative solution of sparse linear systems on graphics processors},

   journal={Computer Science – Research and Development},

   publisher={Springer Berlin / Heidelberg},

   issn={1865-2034},

   keyword={Computer Science},

   pages={1-9},

   url={http://dx.doi.org/10.1007/s00450-011-0195-8},

   note={10.1007/s00450-011-0195-8}

}

Download Download (PDF)   View View   Source Source   

1570

views

In this paper, we analyze the power consumption of different GPU-accelerated iterative solver implementations enhanced with energy-saving techniques. Specifically, while conducting kernel calls on the graphics accelerator, we manually set the host system to a power-efficient idle-wait status so as to leverage dynamic voltage and frequency control. While the usage of iterative refinement combined with mixed precision arithmetic often improves the execution time of an iterative solver on a graphics processor, this may not necessarily be true for the power consumption as well. To analyze the trade-off between computation time and power consumption we compare a plain GMRES solver and its preconditioned variant to the mixed-precision iterative refinement implementations based on the respective solvers. Benchmark experiments conclusively reveal how the usage of idle-wait during GPU-kernel calls effectively leverages the power-tools provided by hardware, and improves the energy performance of the algorithm.
No votes yet.
Please wait...

* * *

* * *

HGPU group © 2010-2024 hgpu.org

All rights belong to the respective authors

Contact us: