Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee

Chayanon Namo Wichitrnithed; Woo Sun Yang; Yun Helen He; Brad Richardson; Koichi Sakaguchi; Manuel Arenaz; William I. Gustafson; Jacob Shpund; Ulises Costi Blanco; Alvaro Goldar Dieste

doi:10.1109/scw63240.2024.00243

Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee

Chayanon Namo Wichitrnithed^*
, Woo Sun Yang
, Yun Helen He
, Brad Richardson
, Koichi Sakaguchi
, Manuel Arenaz
, William I. Gustafson
, Jacob Shpund
, Ulises Costi Blanco
, Alvaro Goldar Dieste

^*Corresponding author for this work

Fredy & Nadine Herrmann Institute of Earth Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Currently, the Weather Research and Forecasting model (WRF) utilizes shared memory (OpenMP) and distributed memory (MPI) parallelisms. To take advantage of GPU resources on the Perlmutter supercomputer at NERSC, we port parts of the computationally expensive routines of the Fast Spectral Bin Microphysics (FSBM) microphysical scheme to NVIDIA GPUs using OpenMP device offloading directives. To facilitate this process, we explore a workflow for optimization which uses both runtime profilers and a static code inspection tool Codee to refactor the subroutine. We observe a 2.08x overall speedup for the CONUS-12km thunderstorm test case.

Original language	English
Title of host publication	Proceedings of SC 2024-W
Subtitle of host publication	Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1934-1942
Number of pages	9
ISBN (Electronic)	9798350355543
DOIs	https://doi.org/10.1109/scw63240.2024.00243
State	Published - 2024
Event	2024 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC Workshops 2024 - Atlanta, United States Duration: 17 Nov 2024 → 22 Nov 2024

Publication series

Name	Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis

Conference

Conference	2024 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC Workshops 2024
Country/Territory	United States
City	Atlanta
Period	17/11/24 → 22/11/24

Bibliographical note

Publisher Copyright:
© 2024 IEEE.

Keywords

Codee
GPU
GPU offloading
Nvidia GPUs
OpenMP
OpenMP Offloading
Weather Research and Forecasting
WRF

Access to Document

10.1109/scw63240.2024.00243

Cite this

Wichitrnithed, C. N., Yang, W. S., He, Y. H., Richardson, B., Sakaguchi, K., Arenaz, M., Gustafson, W. I., Shpund, J., Blanco, U. C., & Dieste, A. G. (2024). Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee. In Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis (pp. 1934-1942). (Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/scw63240.2024.00243

Wichitrnithed, Chayanon Namo ; Yang, Woo Sun ; He, Yun Helen et al. / Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee. Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2024. pp. 1934-1942 (Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis).

@inproceedings{1465d38d6c994e868fe749a58cd27ccf,

title = "Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee",

abstract = "Currently, the Weather Research and Forecasting model (WRF) utilizes shared memory (OpenMP) and distributed memory (MPI) parallelisms. To take advantage of GPU resources on the Perlmutter supercomputer at NERSC, we port parts of the computationally expensive routines of the Fast Spectral Bin Microphysics (FSBM) microphysical scheme to NVIDIA GPUs using OpenMP device offloading directives. To facilitate this process, we explore a workflow for optimization which uses both runtime profilers and a static code inspection tool Codee to refactor the subroutine. We observe a 2.08x overall speedup for the CONUS-12km thunderstorm test case.",

keywords = "Codee, GPU, GPU offloading, Nvidia GPUs, OpenMP, OpenMP Offloading, Weather Research and Forecasting, WRF",

author = "Wichitrnithed, \{Chayanon Namo\} and Yang, \{Woo Sun\} and He, \{Yun Helen\} and Brad Richardson and Koichi Sakaguchi and Manuel Arenaz and Gustafson, \{William I.\} and Jacob Shpund and Blanco, \{Ulises Costi\} and Dieste, \{Alvaro Goldar\}",

note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 2024 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC Workshops 2024 ; Conference date: 17-11-2024 Through 22-11-2024",

year = "2024",

doi = "10.1109/scw63240.2024.00243",

language = "אנגלית",

series = "Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1934--1942",

booktitle = "Proceedings of SC 2024-W",

address = "ארצות הברית",

}

Wichitrnithed, CN, Yang, WS, He, YH, Richardson, B, Sakaguchi, K, Arenaz, M, Gustafson, WI, Shpund, J, Blanco, UC & Dieste, AG 2024, Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee. in Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis. Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis, Institute of Electrical and Electronics Engineers Inc., pp. 1934-1942, 2024 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC Workshops 2024, Atlanta, United States, 17/11/24. https://doi.org/10.1109/scw63240.2024.00243

Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee. / Wichitrnithed, Chayanon Namo; Yang, Woo Sun; He, Yun Helen et al.
Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc., 2024. p. 1934-1942 (Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee

AU - Wichitrnithed, Chayanon Namo

AU - Yang, Woo Sun

AU - He, Yun Helen

AU - Richardson, Brad

AU - Sakaguchi, Koichi

AU - Arenaz, Manuel

AU - Gustafson, William I.

AU - Shpund, Jacob

AU - Blanco, Ulises Costi

AU - Dieste, Alvaro Goldar

PY - 2024

Y1 - 2024

N2 - Currently, the Weather Research and Forecasting model (WRF) utilizes shared memory (OpenMP) and distributed memory (MPI) parallelisms. To take advantage of GPU resources on the Perlmutter supercomputer at NERSC, we port parts of the computationally expensive routines of the Fast Spectral Bin Microphysics (FSBM) microphysical scheme to NVIDIA GPUs using OpenMP device offloading directives. To facilitate this process, we explore a workflow for optimization which uses both runtime profilers and a static code inspection tool Codee to refactor the subroutine. We observe a 2.08x overall speedup for the CONUS-12km thunderstorm test case.

AB - Currently, the Weather Research and Forecasting model (WRF) utilizes shared memory (OpenMP) and distributed memory (MPI) parallelisms. To take advantage of GPU resources on the Perlmutter supercomputer at NERSC, we port parts of the computationally expensive routines of the Fast Spectral Bin Microphysics (FSBM) microphysical scheme to NVIDIA GPUs using OpenMP device offloading directives. To facilitate this process, we explore a workflow for optimization which uses both runtime profilers and a static code inspection tool Codee to refactor the subroutine. We observe a 2.08x overall speedup for the CONUS-12km thunderstorm test case.

KW - Codee

KW - GPU

KW - GPU offloading

KW - Nvidia GPUs

KW - OpenMP

KW - OpenMP Offloading

KW - Weather Research and Forecasting

KW - WRF

UR - https://www.scopus.com/pages/publications/85217178822

U2 - 10.1109/scw63240.2024.00243

DO - 10.1109/scw63240.2024.00243

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85217178822

T3 - Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis

SP - 1934

EP - 1942

BT - Proceedings of SC 2024-W

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2024 Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC Workshops 2024

Y2 - 17 November 2024 through 22 November 2024

ER -

Wichitrnithed CN, Yang WS, He YH, Richardson B, Sakaguchi K, Arenaz M et al. Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee. In Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis. Institute of Electrical and Electronics Engineers Inc. 2024. p. 1934-1942. (Proceedings of SC 2024-W: Workshops of the International Conference for High Performance Computing, Networking, Storage and Analysis). doi: 10.1109/scw63240.2024.00243

Optimizing the Weather Research and Forecasting Model with OpenMP Offload and Codee

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this