Enabling science through emerging HPC technologies: accelerating numerical quadrature using a GPU

Authors

  • I. Spence
  • N.S. Scott
  • C.J. Gillan

Keywords:

numerical integration
R-matrix method
parallel implementation
GPU

Abstract

The R-matrix method when applied to the study of intermediate energy electron scattering by the hydrogen atom gives rise to a large number of two electron integrals between numerical basis functions. Each integral is evaluated independently of the others, thereby rendering this a prime candidate for a parallel implementation. In this paper, we present a parallel implementation of this routine which uses a Graphical Processing Unit as a co-processor, giving a speedup of approximately $20$ times when compared with a sequential version. We briefly consider properties of this calculation which make a GPU implementation appropriate with a view to identifying other calculations which might similarly benefit.

New computational technologies

Downloads

Published

2009-10-18

Issue

Vol. 10 (2009): Issue 4.

Section

Section 1. Numerical methods and applications

Author Biographies

I. Spence

Queen's University of Belfast,
School of Electronics, Electrical Engineering & Computer Science
• Professor

N.S. Scott

Queen's University of Belfast,
School of Electronics, Electrical Engineering & Computer Science
• Professor

C.J. Gillan

Queen's University of Belfast,
School of Electronics, Electrical Engineering & Computer Science
• Professor

References

  1. Parallel linear algebra software for multicore architectures (PLASMA) (http://icl.cs.utk.edu/plasma/, visited 14 September, 2009).
  2. Marc Baboulin, Alfredo Buttari, Jack Dongarra, Jakub Kurzak, Julie Langou, Julien Langou, Piotr Luszczek, and Stanimire Tomov. Accelerating scientific computations with mixed precision algorithms // Computer Physics Communications, In Press, 2008.
    doi 10.1016/j.cpc.2008.11.005
  3. V.M. Burke, C.J. Noble, V. Faro-Maza, A. Maniopoulou, and N.S. Scott. FARM_2DRMP: A version of FARM for use with 2DRMP // Computer Physics Communications, In Press, Accepted Manuscript, 2009. //
    doi 10.1016/j.cpc.2009.07.017
  4. nVidia Corporation. CUBLAS library (http://www.nvidia.com/, visited 18 September, 2009).
  5. nVidia Corporation. nVidia CUDA compute unified device architecture programming guide (version 2.0) // (http://www.nvidia.com/cuda, visited 18 September, 2009).
  6. John D. Owens, Mike Houston, David Luebke, Simon Green, John E. Stone, and James C. Phillips. GPU computing // Proceedings of the IEEE, 96: 879-899, 2008.
  7. Tirath Ramdas, Gregory K. Egan, David Abramson, and Kim K. Baldridge. On ERI sorting for SIMD execution of large-scale hartree-fock SCF // Computer Physics Communications, 178,(11): 817-834, 2008. //
    doi 10.1016/j.cpc.2008.01.045
  8. N.S. Scott, V. Faro-Maza, M.P. Scott, T. Harmer, J.M. Chesneaux, C. Denis, and F. Jézéquel. E-collisions using e-science // Physics of Particles and Nuclei Letters, 5,(3): 150-156, May 2008. DOI:10.1134/S1547477108030023.
  9. N.S. Scott, L.Gr. Ixaru, C. Denis, F. Jézéquel, J.-M. Chesneaux, and M.P. Scott. High performance computation and numerical validation of e-collision software // In G Maroulis and T Simos, editors, Trends and Perspectives in Modern Computational Science, Invited lectures, Vol. 6 of Lecture Series on Computer and Computational Sciences, pages 561-570, 2006.
  10. N.S. Scott, F. Jézéquel, C. Denis, and J.-M. Chesneaux. Numerical `health check» for scientific codes: the CADNA approach // Computer Physics Communications, 176,(8): 507-521, 2007.
    doi 10.1016/j.cpc.2007.01.005
  11. N.S. Scott, M.P. Scott, P.G. Burke, T. Stitt, V. Faro-Maza, C. Denis, and A. Maniopoulou. 2DRMP: A suite of two-dimensional R-matrix propagation codes // Computer Physics Communications, In Press, Accepted Manuscript, 2009.
    doi 10.1016/j.cpc.2009.07.018
  12. ddeke, and Carsten Gutwenger. HONEI: a collection of libraries for numerical computations targeting multiple processor architectures // Computer Physics Communications, In Press, 2009.
    doi 10.1016/j.cpc.2009.04.018