skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Multigroup Monte Carlo on GPUs: Comparison of history- and event-based algorithms

Abstract

This article presents an investigation of the performance of different multigroup Monte Carlo transport algorithms on GPUs with a discussion of both history-based and event-based approaches. Several algorithmic improvements are introduced for both approaches. By modifying the history-based algorithm that is traditionally favored in CPU-based MC codes to occasionally filter out dead particles to reduce thread divergence, performance exceeds that of either the pure history-based or event-based approaches. The impacts of several algorithmic choices are discussed, including performance studies on Kepler and Pascal generation NVIDIA GPUs for fixed source and eigenvalue calculations. Single-device performance equivalent to 20–40 CPU cores on the K40 GPU and 60–80 CPU cores on the P100 GPU is achieved. Last, in addition, nearly perfect multi-device parallel weak scaling is demonstrated on more than 16,000 nodes of the Titan supercomputer.

Authors:
ORCiD logo [1]; ORCiD logo [1]; ORCiD logo [1]
  1. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA); USDOE Office of Science (SC)
OSTI Identifier:
1414707
Grant/Contract Number:
AC05-00OR22725
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Annals of Nuclear Energy (Oxford)
Additional Journal Information:
Journal Name: Annals of Nuclear Energy (Oxford); Journal Volume: 113; Journal Issue: C; Journal ID: ISSN 0306-4549
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Radiation transport; Monte Carlo; GPU

Citation Formats

Hamilton, Steven P., Slattery, Stuart R., and Evans, Thomas M.. Multigroup Monte Carlo on GPUs: Comparison of history- and event-based algorithms. United States: N. p., 2017. Web. doi:10.1016/j.anucene.2017.11.032.
Hamilton, Steven P., Slattery, Stuart R., & Evans, Thomas M.. Multigroup Monte Carlo on GPUs: Comparison of history- and event-based algorithms. United States. doi:10.1016/j.anucene.2017.11.032.
Hamilton, Steven P., Slattery, Stuart R., and Evans, Thomas M.. Fri . "Multigroup Monte Carlo on GPUs: Comparison of history- and event-based algorithms". United States. doi:10.1016/j.anucene.2017.11.032.
@article{osti_1414707,
title = {Multigroup Monte Carlo on GPUs: Comparison of history- and event-based algorithms},
author = {Hamilton, Steven P. and Slattery, Stuart R. and Evans, Thomas M.},
abstractNote = {This article presents an investigation of the performance of different multigroup Monte Carlo transport algorithms on GPUs with a discussion of both history-based and event-based approaches. Several algorithmic improvements are introduced for both approaches. By modifying the history-based algorithm that is traditionally favored in CPU-based MC codes to occasionally filter out dead particles to reduce thread divergence, performance exceeds that of either the pure history-based or event-based approaches. The impacts of several algorithmic choices are discussed, including performance studies on Kepler and Pascal generation NVIDIA GPUs for fixed source and eigenvalue calculations. Single-device performance equivalent to 20–40 CPU cores on the K40 GPU and 60–80 CPU cores on the P100 GPU is achieved. Last, in addition, nearly perfect multi-device parallel weak scaling is demonstrated on more than 16,000 nodes of the Titan supercomputer.},
doi = {10.1016/j.anucene.2017.11.032},
journal = {Annals of Nuclear Energy (Oxford)},
number = C,
volume = 113,
place = {United States},
year = {Fri Dec 22 00:00:00 EST 2017},
month = {Fri Dec 22 00:00:00 EST 2017}
}

Journal Article:
Free Publicly Available Full Text
This content will become publicly available on December 22, 2018
Publisher's Version of Record

Save / Share: