Accelerating time integration for the shallow water equations on the sphere using GPUS

R. Archibald; K. J. Evans; A. Salinger

doi:10.1016/j.procs.2015.05.470

Accelerating time integration for the shallow water equations on the sphere using GPUS

R. Archibald, K. J. Evans, A. Salinger

Research output: Contribution to journal › Conference article › peer-review

2 Scopus citations

Abstract

The push towards larger and larger computational platforms has made it possible for climate simulations to resolve climate dynamics across multiple spatial and temporal scales. This direction in climate simulation has created a strong need to develop scalable time-stepping methods capable of accelerating throughput on high performance computing. This work details the recent advances in the implementation of implicit time stepping on a spectral element cube-sphere grid using graphical processing units (GPU) based machines. We demonstrate how solvers in the Trilinos project are interfaced with ACME and GPU kernels can significantly increase computational speed of the residual calculations in the implicit time stepping method for the shallow water equations on the sphere. We show the optimization gains and data structure reorganization that facilitates the performance improvements.

Original language	English
Pages (from-to)	2046-2055
Number of pages	10
Journal	Procedia Computer Science
Volume	51
Issue number	1
DOIs	https://doi.org/10.1016/j.procs.2015.05.470
State	Published - 2015
Event	International Conference on Computational Science, ICCS 2002 - Amsterdam, Netherlands Duration: Apr 21 2002 → Apr 24 2002

Funding

The submitted manuscript is based upon work, authored in part by contractors [UT-Battelle LLC, manager of Oak Ridge National Laboratory (ORNL)], and supported by the U.S. Department of Energy, Office of Science, Office of BER and ASCR SciDAC project ”Multiscale Methods for Accurate, Efficient, and Scale-Aware Models of the Earth System,” and computed using OLCF computational resources. Accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes. This study used the resources of the Oak Ridge Leadership Computing Facility at the Oak Ridge National Laboratory, which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC05-00OR22725. labelsect:bib

Funders	Funder number
UT-Battelle LLC
U.S. Department of Energy
Office of Science
Advanced Scientific Computing Research
Biological and Environmental Research
Oak Ridge National Laboratory

Keywords

GPU
Implicit timestepping
Trilinos

Access to Document

10.1016/j.procs.2015.05.470

Cite this

@article{13a3b7fc8ce34ff7851b5280f737089a,

title = "Accelerating time integration for the shallow water equations on the sphere using GPUS",

abstract = "The push towards larger and larger computational platforms has made it possible for climate simulations to resolve climate dynamics across multiple spatial and temporal scales. This direction in climate simulation has created a strong need to develop scalable time-stepping methods capable of accelerating throughput on high performance computing. This work details the recent advances in the implementation of implicit time stepping on a spectral element cube-sphere grid using graphical processing units (GPU) based machines. We demonstrate how solvers in the Trilinos project are interfaced with ACME and GPU kernels can significantly increase computational speed of the residual calculations in the implicit time stepping method for the shallow water equations on the sphere. We show the optimization gains and data structure reorganization that facilitates the performance improvements.",

keywords = "GPU, Implicit timestepping, Trilinos",

author = "R. Archibald and Evans, {K. J.} and A. Salinger",

note = "Publisher Copyright: {\textcopyright} The Authors. Published by Elsevier B.V.; International Conference on Computational Science, ICCS 2002 ; Conference date: 21-04-2002 Through 24-04-2002",

year = "2015",

doi = "10.1016/j.procs.2015.05.470",

language = "English",

volume = "51",

pages = "2046--2055",

journal = "Procedia Computer Science",

issn = "1877-0509",

publisher = "Elsevier",

number = "1",

}

TY - JOUR

T1 - Accelerating time integration for the shallow water equations on the sphere using GPUS

AU - Archibald, R.

AU - Evans, K. J.

AU - Salinger, A.

N1 - Publisher Copyright: © The Authors. Published by Elsevier B.V.

PY - 2015

Y1 - 2015

N2 - The push towards larger and larger computational platforms has made it possible for climate simulations to resolve climate dynamics across multiple spatial and temporal scales. This direction in climate simulation has created a strong need to develop scalable time-stepping methods capable of accelerating throughput on high performance computing. This work details the recent advances in the implementation of implicit time stepping on a spectral element cube-sphere grid using graphical processing units (GPU) based machines. We demonstrate how solvers in the Trilinos project are interfaced with ACME and GPU kernels can significantly increase computational speed of the residual calculations in the implicit time stepping method for the shallow water equations on the sphere. We show the optimization gains and data structure reorganization that facilitates the performance improvements.

AB - The push towards larger and larger computational platforms has made it possible for climate simulations to resolve climate dynamics across multiple spatial and temporal scales. This direction in climate simulation has created a strong need to develop scalable time-stepping methods capable of accelerating throughput on high performance computing. This work details the recent advances in the implementation of implicit time stepping on a spectral element cube-sphere grid using graphical processing units (GPU) based machines. We demonstrate how solvers in the Trilinos project are interfaced with ACME and GPU kernels can significantly increase computational speed of the residual calculations in the implicit time stepping method for the shallow water equations on the sphere. We show the optimization gains and data structure reorganization that facilitates the performance improvements.

KW - GPU

KW - Implicit timestepping

KW - Trilinos

UR - http://www.scopus.com/inward/record.url?scp=84939199598&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2015.05.470

DO - 10.1016/j.procs.2015.05.470

M3 - Conference article

AN - SCOPUS:84939199598

SN - 1877-0509

VL - 51

SP - 2046

EP - 2055

JO - Procedia Computer Science

JF - Procedia Computer Science

IS - 1

T2 - International Conference on Computational Science, ICCS 2002

Y2 - 21 April 2002 through 24 April 2002

ER -

Accelerating time integration for the shallow water equations on the sphere using GPUS

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this