Advertisement

Computer Science - Research and Development

, Volume 27, Issue 4, pp 299–307 | Cite as

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

  • Hartwig AnztEmail author
  • Maribel Castillo
  • Juan C. Fernández
  • Vincent Heuveline
  • Francisco D. Igual
  • Rafael Mayo
  • Enrique S. Quintana-Ortí
Special Issue Paper

Abstract

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.

Keywords

Sparse linear systems Iterative solvers GMRES Mixed precision iterative refinement Power-aware algorithms Graphics processors (GPUs) Idle-wait 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Lampe F (2010) Green-IT, Virtualisierung und Thin Clients: Mit neuen IT-Technologien Energieeffizienz erreichen, die Umwelt schonen und Kosten sparen. Vieweg/Teubner, Wiesbaten Google Scholar
  2. 2.
    Kogge P et al (2008) ExaScale computing study: technology challenges in achieving ExaScale systems Google Scholar
  3. 3.
    Dongarra J et al (2011) The international ExaScale software project roadmap. doi: 10.1177/1094342010391989
  4. 4.
    Saad Y (2003) Iterative methods for sparse linear systems. Society for Industrial and Applied Mathematics, Philadelphia zbMATHCrossRefGoogle Scholar
  5. 5.
    Saad Y, Schultz MH (1986) Gmres: a generalized minimal residual algorithm for solving nonsymmetric linear systems. SIAM J Sci Stat Comput 7:856–869. Available: http://portal.acm.org/citation.cfm?id=14063.14074 MathSciNetzbMATHCrossRefGoogle Scholar
  6. 6.
    Anzt H, Rocker B, Heuveline V (2010) Energy efficiency of mixed precision iterative refinement methods using hybrid hardware platforms. Comput Sci Res Dev 25(3):141–149 CrossRefGoogle Scholar
  7. 7.
    Anzt H, Castillo M, Fernández JC, Heuveline V, Mayo R, Quintana-Ortí ES, Rocker B (2011) Power consumption of mixed precision in the iterative solution of sparse linear systems. In: Proceedings of the 25th IEEE international parallel & distributed processing symposium Google Scholar
  8. 8.
    Anzt H, Aliaga J, Castillo M, Fernández JC, Mayo R, Quintana-Ortí ES (2011) Analysis and optimization of power consumption in the iterative solution of sparse linear systems on multi-core and many-core platforms. EMCL Preprint Series. Available: http://www.emcl.kit.edu/preprints/emcl-preprint-2011-05.pdf
  9. 9.
    NVIDIA CUDA CUBLAS library programming guide, 1st edn, NVIDIA Corporation (June 2007) Google Scholar
  10. 10.
    Bell N, Garland M (2009) Implementing sparse matrix-vector multiplication on throughput-oriented processors. In: Proceedings of the conference on high performance computing networking, storage and analysis, SC’09. ACM, New York, pp 18:1–18:11 Google Scholar
  11. 11.
    Palma J, Daydé M, Marques O, Lopes J (eds) (2011) An error correction solver for linear systems: evaluation of mixed precision implementations. Lecture notes in computer science, vol 6449. Springer, Berlin Google Scholar
  12. 12.
    Dongarra JJ, Duff IS, Sorensen DC, van der Vorst HA (1998) Numerical linear algebra for high-performance computers. Society for Industrial and Applied Mathematics, Philadelphia zbMATHCrossRefGoogle Scholar
  13. 13.
    Göddeke D, Strzodka R, Turek S (2007) Performance and accuracy of hardware-oriented native-, emulated- and mixed-precision solvers in FEM simulations. Int J Parallel Emergent Distributed Syst 22(4):221–256 zbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  • Hartwig Anzt
    • 1
    Email author
  • Maribel Castillo
    • 2
  • Juan C. Fernández
    • 2
  • Vincent Heuveline
    • 1
  • Francisco D. Igual
    • 2
  • Rafael Mayo
    • 2
  • Enrique S. Quintana-Ortí
    • 2
  1. 1.Institute for Applied and Numerical Mathematics 4Karlsruhe Institute of TechnologyKarlsruheGermany
  2. 2.Depto. de Ingeniería y Ciencia de ComputadoresUniversidad Jaume ICastellónSpain

Personalised recommendations