Colloquium: Large scale simulations on GPU clusters

  • Massimo Bernaschi
  • Mauro Bisson
  • Massimiliano Fatica
Colloquium

DOI: 10.1140/epjb/e2015-60180-8

Cite this article as:
Bernaschi, M., Bisson, M. & Fatica, M. Eur. Phys. J. B (2015) 88: 158. doi:10.1140/epjb/e2015-60180-8

Abstract

Graphics processing units (GPU) are currently used as a cost-effective platform for computer simulations and big-data processing. Large scale applications require that multiple GPUs work together but the efficiency obtained with cluster of GPUs is, at times, sub-optimal because the GPU features are not exploited at their best. We describe how it is possible to achieve an excellent efficiency for applications in statistical mechanics, particle dynamics and networks analysis by using suitable memory access patterns and mechanisms like CUDA streams, profiling tools, etc. Similar concepts and techniques may be applied also to other problems like the solution of Partial Differential Equations.

Keywords

Statistical and Nonlinear Physics 

Copyright information

© EDP Sciences, SIF, Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • Massimo Bernaschi
    • 1
  • Mauro Bisson
    • 2
  • Massimiliano Fatica
    • 2
  1. 1.Istituto per le Applicazioni del Calcolo, National Research Council of ItalyRomaItaly
  2. 2.NVIDIA Corporation, 2701 San Tomas ExpresswaySanta ClaraUSA

Personalised recommendations