Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive‐Based Methods
Abstract Cloud microphysics is one of the most time‐consuming components in a climate model. In this study, we port the cloud microphysics parameterization in the Community Atmosphere Model (CAM), known as Parameterization of Unified Microphysics Across Scales (PUMAS), from CPU to GPU to seek a comp...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
American Geophysical Union (AGU)
2023-05-01
|
| Series: | Journal of Advances in Modeling Earth Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1029/2022MS003515 |
| _version_ | 1827921691289845760 |
|---|---|
| author | Jian Sun John M. Dennis Sheri A. Mickelson Brian Vanderwende Andrew Gettelman Katherine Thayer‐Calder |
| author_facet | Jian Sun John M. Dennis Sheri A. Mickelson Brian Vanderwende Andrew Gettelman Katherine Thayer‐Calder |
| author_sort | Jian Sun |
| collection | DOAJ |
| description | Abstract Cloud microphysics is one of the most time‐consuming components in a climate model. In this study, we port the cloud microphysics parameterization in the Community Atmosphere Model (CAM), known as Parameterization of Unified Microphysics Across Scales (PUMAS), from CPU to GPU to seek a computational speedup. The directive‐based methods (OpenACC and OpenMP target offload) are determined as the best fit specifically for our development practices, which enable a single version of source code to run either on the CPU or GPU, and yield a better portability and maintainability. Their performance is first examined in a PUMAS stand‐alone kernel and the directive‐based methods can outperform a CPU node as long as there is enough computational burden on the GPU. A consistent behavior is observed when we run PUMAS on the GPU in a practical CAM simulation. A 3.6× speedup of the PUMAS execution time, including data movement between CPU and GPU, is achieved at a coarse horizontal resolution (8 NVIDIA V100 GPUs against 36 Intel Skylake CPU cores). This speedup further increases up to 5.4× at a high resolution (24 NVIDIA V100 GPUs against 108 Intel Skylake CPU cores), which highlights the fact that GPU favors larger problem size. This study demonstrates that using GPU in a CAM simulation can save noticeable computational costs even with a small portion of code being GPU‐enabled. Therefore, we are encouraged to port more parameterizations to GPU to take advantage of its computational benefit. |
| first_indexed | 2024-03-13T04:31:04Z |
| format | Article |
| id | doaj.art-81895c2ebdef4b69b2d37ff4e66334c7 |
| institution | Directory Open Access Journal |
| issn | 1942-2466 |
| language | English |
| last_indexed | 2024-03-13T04:31:04Z |
| publishDate | 2023-05-01 |
| publisher | American Geophysical Union (AGU) |
| record_format | Article |
| series | Journal of Advances in Modeling Earth Systems |
| spelling | doaj.art-81895c2ebdef4b69b2d37ff4e66334c72023-06-19T13:40:46ZengAmerican Geophysical Union (AGU)Journal of Advances in Modeling Earth Systems1942-24662023-05-01155n/an/a10.1029/2022MS003515Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive‐Based MethodsJian Sun0John M. Dennis1Sheri A. Mickelson2Brian Vanderwende3Andrew Gettelman4Katherine Thayer‐Calder5National Center for Atmospheric Research CO Boulder USANational Center for Atmospheric Research CO Boulder USANational Center for Atmospheric Research CO Boulder USANational Center for Atmospheric Research CO Boulder USANational Center for Atmospheric Research CO Boulder USANational Center for Atmospheric Research CO Boulder USAAbstract Cloud microphysics is one of the most time‐consuming components in a climate model. In this study, we port the cloud microphysics parameterization in the Community Atmosphere Model (CAM), known as Parameterization of Unified Microphysics Across Scales (PUMAS), from CPU to GPU to seek a computational speedup. The directive‐based methods (OpenACC and OpenMP target offload) are determined as the best fit specifically for our development practices, which enable a single version of source code to run either on the CPU or GPU, and yield a better portability and maintainability. Their performance is first examined in a PUMAS stand‐alone kernel and the directive‐based methods can outperform a CPU node as long as there is enough computational burden on the GPU. A consistent behavior is observed when we run PUMAS on the GPU in a practical CAM simulation. A 3.6× speedup of the PUMAS execution time, including data movement between CPU and GPU, is achieved at a coarse horizontal resolution (8 NVIDIA V100 GPUs against 36 Intel Skylake CPU cores). This speedup further increases up to 5.4× at a high resolution (24 NVIDIA V100 GPUs against 108 Intel Skylake CPU cores), which highlights the fact that GPU favors larger problem size. This study demonstrates that using GPU in a CAM simulation can save noticeable computational costs even with a small portion of code being GPU‐enabled. Therefore, we are encouraged to port more parameterizations to GPU to take advantage of its computational benefit.https://doi.org/10.1029/2022MS003515GPUOpenACCOpenMP target offloadPUMASCAM |
| spellingShingle | Jian Sun John M. Dennis Sheri A. Mickelson Brian Vanderwende Andrew Gettelman Katherine Thayer‐Calder Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive‐Based Methods Journal of Advances in Modeling Earth Systems GPU OpenACC OpenMP target offload PUMAS CAM |
| title | Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive‐Based Methods |
| title_full | Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive‐Based Methods |
| title_fullStr | Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive‐Based Methods |
| title_full_unstemmed | Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive‐Based Methods |
| title_short | Acceleration of the Parameterization of Unified Microphysics Across Scales (PUMAS) on the Graphics Processing Unit (GPU) With Directive‐Based Methods |
| title_sort | acceleration of the parameterization of unified microphysics across scales pumas on the graphics processing unit gpu with directive based methods |
| topic | GPU OpenACC OpenMP target offload PUMAS CAM |
| url | https://doi.org/10.1029/2022MS003515 |
| work_keys_str_mv | AT jiansun accelerationoftheparameterizationofunifiedmicrophysicsacrossscalespumasonthegraphicsprocessingunitgpuwithdirectivebasedmethods AT johnmdennis accelerationoftheparameterizationofunifiedmicrophysicsacrossscalespumasonthegraphicsprocessingunitgpuwithdirectivebasedmethods AT sheriamickelson accelerationoftheparameterizationofunifiedmicrophysicsacrossscalespumasonthegraphicsprocessingunitgpuwithdirectivebasedmethods AT brianvanderwende accelerationoftheparameterizationofunifiedmicrophysicsacrossscalespumasonthegraphicsprocessingunitgpuwithdirectivebasedmethods AT andrewgettelman accelerationoftheparameterizationofunifiedmicrophysicsacrossscalespumasonthegraphicsprocessingunitgpuwithdirectivebasedmethods AT katherinethayercalder accelerationoftheparameterizationofunifiedmicrophysicsacrossscalespumasonthegraphicsprocessingunitgpuwithdirectivebasedmethods |