Implementing a GPU-portable field line tracing application with OpenMP Offload

Jiménez, Diego; Herrera-Mora, Javier; Rampp, Markus; Laure, Erwin; Meneses, Esteban

doi:10.1007/978-3-031-23821-5_3

Item

ITEM ACTIONSEXPORT

Add to Basket

Local TagsRelease HistoryDetailsSummary

Released

Conference Paper

Implementing a GPU-portable field line tracing application with OpenMP Offload

MPS-Authors

/persons/resource/persons110221

Rampp, Markus
Max Planck Computing and Data Facility, Max Planck Society;

/persons/resource/persons264425

Laure, Erwin
Max Planck Computing and Data Facility, Max Planck Society;

External Resource

No external resources are shared

Fulltext (restricted access)

There are currently no full texts shared for your IP range.

Fulltext (public)

There are no public fulltexts stored in PuRe

Supplementary Material (public)

There is no public supplementary material available

Citation

Jiménez, D., Herrera-Mora, J., Rampp, M., Laure, E., & Meneses, E. (2022). Implementing a GPU-portable field line tracing application with OpenMP Offload. In P. Navaux, C. J. Barrios H., C. Osthoff, & G. Guerrero (Eds.), High Performance Computing: 9th Latin American Conference, CARLA 2022, Porto Alegre, Brazil, September 26–30, 2022, Revised Selected Papers (pp. 31-46). Midtown Manhattan, New York City: Springer Cham. doi:10.1007/978-3-031-23821-5_3.

Cite as: https://hdl.handle.net/21.11116/0000-000C-1E1D-4

Abstract

Accelerated computing is becoming more diverse as new vendors and architectures come into play. Although platform-specific programming models promise ease of development and better control over performance, they still restrict the portability of scientific applications. As the OpenMP offloading specification becomes adopted by more compilers, this programming model stands out as a vendor-neutral portable approach to heterogeneous programming. In this study, we port a plasma physics oriented field line tracing code from a CPU-based MPI+OpenMP approach to a GPU accelerated version, using OpenMP’s offloading capabilities. We analyze GPU performance across different vendors with respect to the original CPU version and test both prescriptive and descriptive approaches to accelerator programming. A maximum 6× acceleration over the CPU implementation was achieved using OpenMP’s high-level offloading directives. In addition, we demonstrate portability across three different vendor GPUs with no code modifications.