arrow
返回

Pace v0.2: a Python-based performance-portable atmospheric model

delete2023-05-17
delete3
delete
OA
AI
J
Johann Dahm
E
Eddie Davis
F
Florian Deconinck
O
Oliver D. Elbert
R
Rhea George
J
Jeremy McGibbon
T
Tobias Wicky
E
Elynn Wu
C
Christopher Kung
T
Tal Ben‐Nun
L
Lucas Harris
L
Linus Groner
O
Oliver Fuhrer *
DOI:10.5194/gmd-16-2719-2023delete
delete原文链接
delete分享
delete收藏
摘要

摘要

En 中文
Progress in leveraging current and emerging high-performance computing infrastructures using traditional weather and climate models has been slow. This has become known more broadly as the software productivity gap. With the end of Moore's law driving forward rapid specialization of hardware architectures, building simulation codes on a low-level language with hardware-specific optimizations is a significant risk. As a solution, we present Pace, an implementation of the nonhydrostatic FV3 dynamical core and GFDL cloud microphysics scheme which is entirely Python-based. In order to achieve high performance on a diverse set of hardware architectures, Pace is written using the GT4Py domain-specific language. We demonstrate that with this approach we can achieve portability and performance, while significantly improving the readability and maintainability of the code as compared to the Fortran reference implementation. We show that Pace can run at scale on leadership-class supercomputers and achieve performance speeds 3.5-4 times faster than the Fortran code on GPU-accelerated supercomputers. Furthermore, we demonstrate how a Python-based simulation code facilitates existing or enables entirely new use cases and workflows. Pace demonstrates how a high-level language can insulate us from disruptive changes, provide a more productive development environment, and facilitate the integration with new technologies such as machine learning.
Keyword:
DYNAMICAL CORE
TRANSPORT
GPUS

期刊

Geoscientific Model Development 封面图
Geoscientific Model Development
IF:
4.9
论文数:
4.1K
被引数:
3.4W

机构

E
eth zurich
学者数:
4.2K
论文数: 1.4K
被引数: 3.2W
N
national aeronautics & space administration (nasa)
学者数:
3.1W
论文数: 2.6W
被引数: 3.7W
N
national oceanic atmospheric admin (noaa) - usa
学者数:
0
论文数: 226
被引数: 1.1W
N
nasa goddard space flight center
学者数:
470
论文数: 237
被引数: 8.4K
S
swiss federal institutes of technology domain
学者数:
795
论文数: 1.1K
被引数: 9.3W
学者 查看更多机构
引用论文

引用论文

Dedalus: A flexible framework for numerical simulations with spectral methodsDedalus: 用谱方法进行数值模拟的灵活框架
err2020-04-23
err287
errOAAI
errBurns, Keaton J.; Vasil, Geoffrey M.; Oishi, Jeffrey S.; Lecoanet, Daniel; Brown, Benjamin P.
err分享
err收藏
The resource theory of stabilizer quantum computation
err2014-01-09
err0
errOAAI
errVictor Veitch; S A Hamed Mousavian; Daniel Gottesman; Joseph Emerson
err分享
err收藏
学者 查看更多内容