Biological and Environmental Research - Earth and Environmental System Sciences
Earth and Environmental System Modeling

Accelerating Time Integration for the Shallow Water Equations on the Sphere Using GPUs

TitleAccelerating Time Integration for the Shallow Water Equations on the Sphere Using GPUs
Publication TypeJournal Article
Year of Publication2015
AuthorsArchibald, Richard, Evans Katherine, and Salinger Andrew
JournalProcedia Computer Science
Volume51
Pages2046-2055
Abstract / Summary

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 that 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.

URLhttp://dx.doi.org/10.1016/j.procs.2015.05.470
DOI10.1016/j.procs.2015.05.470
Journal: Procedia Computer Science
Year of Publication: 2015
Volume: 51
Pages: 2046-2055
Publication Date: 06/2015

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 that 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.

DOI: 10.1016/j.procs.2015.05.470
Citation:
Archibald, R, K Evans, and A Salinger.  2015.  "Accelerating Time Integration for the Shallow Water Equations on the Sphere Using GPUs."  Procedia Computer Science 51: 2046-2055.  https://doi.org/10.1016/j.procs.2015.05.470.