Pseudorandom Numbers Generation for Monte Carlo Simulations on GPUs: OpenCL Approach
Dnipropetrovsk National University, Dnipropetrovsk, Ukraine
Chapter 12 in book "Numerical Computations with GPUs", 2014
@incollection{demchik2014pseudorandom,
title={Pseudorandom Numbers Generation for Monte Carlo Simulations on GPUs: OpenCL Approach},
author={Demchik, Vadim},
booktitle={Numerical Computations with GPUs},
pages={245–271},
year={2014},
publisher={Springer}
}
General principles of pseudorandom numbers production for Monte Carlo simulations on GPUs are discussed by creating an OpenCL open-source library of pseudorandom number generators PRNGCL. The library contains implementation of a number of the most popular uniform generators. The most popular pseudorandom number generators for Monte Carlo simulations and libraries for GPUs are reviewed. Some tricks for generators implementation on GPUs with OpenCL framework are considered. The performance results of implemented generators on several AMD and NVIDIA GPUs are presented.
March 25, 2015 by hgpu
