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

Title: Demonstrating GPU code portability and scalability for radiative heat transfer computations

Abstract

High performance computing frameworks utilizing CPUs, Nvidia GPUs, and/or Intel Xeon Phis necessitate portable and scalable solutions for application developers. Nvidia GPUs in particular present numerous portability challenges with a different programming model, additional memory hierarchies, and partitioned execution units among streaming multiprocessors. Here, this work presents modifications to the Uintah asynchronous many-task runtime and the Kokkos portability library to enable one single codebase for complex multiphysics applications to run across different architectures. Scalability and performance results are shown on multiple architectures for a globally coupled radiation heat transfer simulation, ranging from a single node to 16384 Titan compute nodes.

Authors:
 [1];  [1];  [1];  [1];  [1];  [2];  [3]
  1. Univ. of Utah, Salt Lake City, UT (United States)
  2. Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
  3. NVIDIA Corp., Santa Clara, CA (United States)
Publication Date:
Research Org.:
Univ. of Utah, Salt Lake City, UT (United States)
Sponsoring Org.:
USDOE National Nuclear Security Administration (NNSA)
OSTI Identifier:
1582435
Alternate Identifier(s):
OSTI ID: 1694247
Grant/Contract Number:  
NA0002375; AC05-00OR22725; AC02-06CH11357
Resource Type:
Accepted Manuscript
Journal Name:
Journal of Computational Science
Additional Journal Information:
Journal Volume: 27; Journal Issue: C; Journal ID: ISSN 1877-7503
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; Asynchronous Many-Task Runtime; GPU; Scalability; Portability; Radiative Heat Transfer

Citation Formats

Peterson, Brad, Humphrey, Alan, Holmen, John, Harman, Todd, Berzins, Martin, Sunderland, Dan, and Edwards, H. Carter. Demonstrating GPU code portability and scalability for radiative heat transfer computations. United States: N. p., 2018. Web. https://doi.org/10.1016/j.jocs.2018.06.005.
Peterson, Brad, Humphrey, Alan, Holmen, John, Harman, Todd, Berzins, Martin, Sunderland, Dan, & Edwards, H. Carter. Demonstrating GPU code portability and scalability for radiative heat transfer computations. United States. https://doi.org/10.1016/j.jocs.2018.06.005
Peterson, Brad, Humphrey, Alan, Holmen, John, Harman, Todd, Berzins, Martin, Sunderland, Dan, and Edwards, H. Carter. Fri . "Demonstrating GPU code portability and scalability for radiative heat transfer computations". United States. https://doi.org/10.1016/j.jocs.2018.06.005. https://www.osti.gov/servlets/purl/1582435.
@article{osti_1582435,
title = {Demonstrating GPU code portability and scalability for radiative heat transfer computations},
author = {Peterson, Brad and Humphrey, Alan and Holmen, John and Harman, Todd and Berzins, Martin and Sunderland, Dan and Edwards, H. Carter},
abstractNote = {High performance computing frameworks utilizing CPUs, Nvidia GPUs, and/or Intel Xeon Phis necessitate portable and scalable solutions for application developers. Nvidia GPUs in particular present numerous portability challenges with a different programming model, additional memory hierarchies, and partitioned execution units among streaming multiprocessors. Here, this work presents modifications to the Uintah asynchronous many-task runtime and the Kokkos portability library to enable one single codebase for complex multiphysics applications to run across different architectures. Scalability and performance results are shown on multiple architectures for a globally coupled radiation heat transfer simulation, ranging from a single node to 16384 Titan compute nodes.},
doi = {10.1016/j.jocs.2018.06.005},
journal = {Journal of Computational Science},
number = C,
volume = 27,
place = {United States},
year = {2018},
month = {6}
}

Journal Article:

Citation Metrics:
Cited by: 6 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Radiative Heat Transfer Calculation on 16384 GPUs Using a Reverse Monte Carlo Ray Tracing Approach with Adaptive Mesh Refinement
conference, May 2016

  • Humphrey, Alan; Sunderland, Daniel; Harman, Todd
  • 2016 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW)
  • DOI: 10.1109/IPDPSW.2016.93

Application of a particle-in-cell method to solid mechanics
journal, May 1995

  • Sulsky, Deborah; Zhou, Shi-Jian; Schreyer, Howard L.
  • Computer Physics Communications, Vol. 87, Issue 1-2
  • DOI: 10.1016/0010-4655(94)00170-7

Scalable large-scale fluid-structure interaction solvers in the Uintah framework via hybrid task-based parallelism algorithms: UINTAH HYBRID TASK-BASED PARALLELISM ALGORITHMS
journal, July 2013

  • Meng, Qingyu; Berzins, Martin
  • Concurrency and Computation: Practice and Experience, Vol. 26, Issue 7
  • DOI: 10.1002/cpe.3099

Large Eddy Simulations of Accidental Fires Using Massively Parallel Computers
conference, June 2012

  • Smith, Philip; Rawat, Rajesh; Spinti, Jennifer
  • 16th AIAA Computational Fluid Dynamics Conference
  • DOI: 10.2514/6.2003-3697

Large Eddy Simulation of Pulverized Coal Jet Flame Ignition Using the Direct Quadrature Method of Moments
journal, October 2012

  • Pedel, Julien; Thornock, Jeremy N.; Smith, Philip J.
  • Energy & Fuels, Vol. 26, Issue 11
  • DOI: 10.1021/ef3012905

Strong Stability-Preserving High-Order Time Discretization Methods
journal, January 2001


Parallel Computations of Radiative heat Transfer Using the Discrete Ordinates Method
journal, December 2004

  • Krishnamoorthy, Gautham; Rawat, Rajesh; Smith, Philip J.
  • Numerical Heat Transfer, Part B: Fundamentals, Vol. 47, Issue 1
  • DOI: 10.1080/10407790490487451

Parallelization of the P-1 Radiation Model
journal, January 2006

  • Krishnamoorthy, Gautham; Rawat, Rajesh; Smith, Philip J.
  • Numerical Heat Transfer, Part B: Fundamentals, Vol. 49, Issue 1
  • DOI: 10.1080/10407790500344068

A Parametric Case Study in Radiative Heat Transfer Using the Reverse Monte-Carlo Ray-Tracing With Full-Spectrum k-Distribution Method
journal, December 2009

  • Sun, Xiaojing; Smith, Philip J.
  • Journal of Heat Transfer, Vol. 132, Issue 2
  • DOI: 10.1115/1.4000232

Spatial Domain-Based Parallelism in Large-Scale, Participating-Media, Radiative Transport Applications
journal, June 1997

  • Burns, Shawn P.; Christen, Mark A.
  • Numerical Heat Transfer, Part B: Fundamentals, Vol. 31, Issue 4
  • DOI: 10.1080/10407799708915117

Kokkos: Enabling manycore performance portability through polymorphic memory access patterns
journal, December 2014

  • Carter Edwards, H.; Trott, Christian R.; Sunderland, Daniel
  • Journal of Parallel and Distributed Computing, Vol. 74, Issue 12
  • DOI: 10.1016/j.jpdc.2014.07.003

OpenMP: an industry standard API for shared-memory programming
journal, January 1998

  • Dagum, L.; Menon, R.
  • IEEE Computational Science and Engineering, Vol. 5, Issue 1
  • DOI: 10.1109/99.660313

FLASH: An Adaptive Mesh Hydrodynamics Code for Modeling Astrophysical Thermonuclear Flashes
journal, November 2000

  • Fryxell, B.; Olson, K.; Ricker, P.
  • The Astrophysical Journal Supplement Series, Vol. 131, Issue 1
  • DOI: 10.1086/317361

Crash: a Block-Adaptive-Mesh code for Radiative Shock Hydrodynamics—Implementation and Verification
journal, May 2011

  • van der Holst, B.; Tóth, G.; Sokolov, I. V.
  • The Astrophysical Journal Supplement Series, Vol. 194, Issue 2
  • DOI: 10.1088/0067-0049/194/2/23

An Adaptive Mesh Refinement Algorithm for the Radiative Transport Equation
journal, January 1998

  • Jessee, J. Patrick; Fiveland, Woodrow A.; Howell, Louis H.
  • Journal of Computational Physics, Vol. 139, Issue 2
  • DOI: 10.1006/jcph.1997.5870

Solution of Equilibrium Radiation Diffusion Problems Using Implicit Adaptive Mesh Refinement
journal, January 2006

  • Pernice, Michael; Philip, Bobby
  • SIAM Journal on Scientific Computing, Vol. 27, Issue 5
  • DOI: 10.1137/040609069

DAGuE: A generic distributed DAG engine for High Performance Computing
journal, January 2012


StarPU: a unified platform for task scheduling on heterogeneous multicore architectures
journal, November 2010

  • Augonnet, Cédric; Thibault, Samuel; Namyst, Raymond
  • Concurrency and Computation: Practice and Experience, Vol. 23, Issue 2
  • DOI: 10.1002/cpe.1631

    Works referencing / citing this record:

    swHPFM: Refactoring and Optimizing the Structured Grid Fluid Mechanical Algorithm on the Sunway TaihuLight Supercomputer
    journal, December 2019

    • Li, Jingbo; Zhang, Xingjun; Zhou, Jianfeng
    • Applied Sciences, Vol. 10, Issue 1
    • DOI: 10.3390/app10010072

    Structured mesh-oriented framework design and optimization for a coarse-grained parallel CFD solver based on hybrid MPI/OpenMP programming
    journal, November 2019