the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Exploring a high-level programming model for the NWP domain using ECMWF microphysics schemes
Abstract. We explore the domain-specific Python library GT4Py (GridTools for Python) for implementing a representative physical parametrization scheme and the related tangent-linear and adjoint algorithms from the Integrated Forecasting System (IFS) of ECMWF. GT4Py encodes stencil operators in an abstract and hardware-agnostic fashion, thus enabling more concise, readable and maintainable scientific applications. The library achieves high performance by translating the application into targeted low-level coding implementations. Here, the main goal is to study the correctness and performance-portability of the Python rewrites with GT4Py against the reference Fortran code and a number of automatically and manually ported variants created by ECMWF. The present work is part of a larger cross-institutional effort to port weather and climate models to Python with GT4Py. The focus of the current work is the IFS prognostic cloud microphysics scheme, a core physical parametrization represented by a comprehensive code that takes a significant share of the total forecast model execution time. In order to verify GT4Py for Numerical Weather Prediction (NWP) systems, we put additional emphasis on the implementation and validation of the tangent-linear and adjoint model versions which are employed in data assimilation. We benchmark all prototype codes on three European supercomputers characterized by diverse GPU and CPU hardware, node designs, software stacks and compiler suites. Once the application is ported to Python with GT4Py, we find excellent portability, competitive performance, and robust execution in all tested scenarios including with reduced precision.
- Preprint
(1148 KB) - Metadata XML
- BibTeX
- EndNote
Status: open (until 29 Jul 2024)
-
RC1: 'Comment on gmd-2024-92', Anonymous Referee #1, 25 Jun 2024
reply
As a developer who is using GT4Py to port parameterized physics, I am encouraged by these performance results as well as the portability across multiple GPU architectures. Overall, I think this is an excellent paper that highlights the potential of DSLs as a forward-looking development platform. I have several questions and comments.
- Line 191 : This line mentions that " can be differentiated for the vertical boundaries using the interval context manager". As a GT4Py user, it's clear what is being written, but given that "differentiated" has mathematical meanings, it may be better to reword this to avoid confusion.
- List 1 and 2 : I realized later that the "Code and data availability" section lists the repositories that contain the codes in List 1 and 2. Originally, I had mistakenly searched the ECMWF-iFS Github site for the CLOUDSC and CLOUDSC2 dwarf codes and was wondering why I couldn't find the codes from the list. One suggestion is to mention that the repos for the codes are mentioned later in the "Code and data availablility" section.
- Line 296 : Can NPROMA be explained further?
- Line 307 : To clarify, is the symmetry test timing the sum of the CLOUDSC2TL and CLOUDSCAD timings?
- Line 336 : I'm a bit confused on the virtual GPU explanation. Does this mean that when 1 MPI process is mapped to an MI250X, only half the GPU is executed?
- Question: The Gridtools backend was mentioned as a GT4Py backend (and I think it enables GPU compute), but its results were not presented. Was it because it was slower than the Dace backend?
Citation: https://doi.org/10.5194/gmd-2024-92-RC1 -
RC2: 'Comment on gmd-2024-92', Anonymous Referee #2, 26 Jun 2024
reply
The comment was uploaded in the form of a supplement: https://gmd.copernicus.org/preprints/gmd-2024-92/gmd-2024-92-RC2-supplement.pdf
Viewed
HTML | XML | Total | BibTeX | EndNote | |
---|---|---|---|---|---|
217 | 65 | 14 | 296 | 7 | 4 |
- HTML: 217
- PDF: 65
- XML: 14
- Total: 296
- BibTeX: 7
- EndNote: 4
Viewed (geographical distribution)
Country | # | Views | % |
---|
Total: | 0 |
HTML: | 0 |
PDF: | 0 |
XML: | 0 |
- 1