Advertisement

Accelerating VASP electronic structure calculations using graphic processing units

Authors

  • Mohamed Hacene,

    1. IFP Energies Nouvelles, 1 et 4 avenue de Bois-Préau, F-92852 Rueil-Malmaison Cedex, France
    Search for more papers by this author
  • Ani Anciaux-Sedrakian,

    1. IFP Energies Nouvelles, 1 et 4 avenue de Bois-Préau, F-92852 Rueil-Malmaison Cedex, France
    Search for more papers by this author
  • Xavier Rozanska,

    1. Laboratoire de Chimie de l'ENS de Lyon, Université de Lyon, UMR CNRS 5182, 46 Allée d'Italie, F-69364 Lyon Cedex 07, France
    Current affiliation:
    1. Materials Design, 18 rue de Saisset, F-92120 Montrouge, France
    Search for more papers by this author
  • Diego Klahr,

    1. IFP Energies Nouvelles, 1 et 4 avenue de Bois-Préau, F-92852 Rueil-Malmaison Cedex, France
    Current affiliation:
    1. Total E&P, Centre Scientifique et Technique J. Feger, Avenue Larribau, F-64000 Pau, France
    Search for more papers by this author
  • Thomas Guignon,

    Corresponding author
    1. IFP Energies Nouvelles, 1 et 4 avenue de Bois-Préau, F-92852 Rueil-Malmaison Cedex, France
    • IFP Energies Nouvelles, 1 et 4 avenue de Bois-Préau, F-92852 Rueil-Malmaison Cedex, France
    Search for more papers by this author
  • Paul Fleurat-Lessard

    Corresponding author
    1. Laboratoire de Chimie de l'ENS de Lyon, Université de Lyon, UMR CNRS 5182, 46 Allée d'Italie, F-69364 Lyon Cedex 07, France
    • IFP Energies Nouvelles, 1 et 4 avenue de Bois-Préau, F-92852 Rueil-Malmaison Cedex, France
    Search for more papers by this author

Abstract

We present a way to improve the performance of the electronic structure Vienna Ab initio Simulation Package (VASP) program. We show that high-performance computers equipped with graphics processing units (GPUs) as accelerators may reduce drastically the computation time when offloading these sections to the graphic chips. The procedure consists of (i) profiling the performance of the code to isolate the time-consuming parts, (ii) rewriting these so that the algorithms become better-suited for the chosen graphic accelerator, and (iii) optimizing memory traffic between the host computer and the GPU accelerator. We chose to accelerate VASP with NVIDIA GPU using CUDA. We compare the GPU and original versions of VASP by evaluating the Davidson and RMM-DIIS algorithms on chemical systems of up to 1100 atoms. In these tests, the total time is reduced by a factor between 3 and 8 when running on n (CPU core + GPU) compared to n CPU cores only, without any accuracy loss. © 2012 Wiley Periodicals, Inc.

Ancillary