Articles | Volume 16, issue 9
Development and technical paper
 | Highlight paper
17 May 2023
Development and technical paper | Highlight paper |  | 17 May 2023

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

Johann Dahm, Eddie Davis, Florian Deconinck, Oliver Elbert, Rhea George, Jeremy McGibbon, Tobias Wicky, Elynn Wu, Christopher Kung, Tal Ben-Nun, Lucas Harris, Linus Groner, and Oliver Fuhrer

Model code and software

ai2cm/pace: v0.2.0 Rhea George, Elynn Wu, Jeremy McGibbon, Johann Dahm, Eddie Davis, Tobias Wicky, Florian Deconinck, Christopher Kung, Oliver Fuhrer, Oliver Elbert, Ajda Savarin, Noah D. Brenowitz, Mark Cheeseman, Brian Henn, Spencer Clark, and Yannick Niedermayr

ai2cm/gt4py: v0.1.0 GMD release Johann Dahm, Linus Groner, Enrique G. Paredes, Felix Thaler, Hannes Vogt, Eddie Davis, Rico Haeuselmann, Till Ehrengruber, Stefano Ubbiali, Tobias Wicky, Florian Deconinck, Tal Ben-Nun, and Rhea George

Executive editor
Achieving both performance and portability in a whole dynamical core implemented in a high-productivity language such as Python is an eye-opening result which rebuts some widely held assumptions in the geoscientific modelling community. This is a paper which everyone who writes geoscientific models should read.
Short summary
It is hard for scientists to write code which is efficient on different kinds of supercomputers. Python is popular for its user-friendliness. We converted a Fortran code, simulating Earth's atmosphere, into Python. This new code auto-converts to a faster language for processors or graphic cards. Our code runs 3.5–4 times faster on graphic cards than the original on processors in a specific supercomputer system.