Advanced search
Start date
Betweenand

Time parallelism in Open-IFS

Grant number: 23/04579-5
Support Opportunities:Scholarships in Brazil - Post-Doctoral
Effective date (Start): July 01, 2023
Effective date (End): June 30, 2025
Field of knowledge:Physical Sciences and Mathematics - Mathematics - Applied Mathematics
Principal Investigator:Pedro da Silva Peixoto
Grantee:Guilherme Luiz Torres Mendonça
Host Institution: Instituto de Matemática e Estatística (IME). Universidade de São Paulo (USP). São Paulo , SP, Brazil
Associated research grant:21/06176-0 - Numerical methods for a new generation of weather and climate models, AP.PFPMCG.JP2

Abstract

This project aims at exploring the parallel-in-time possibilities within OpenIFS, an open-source version of one of the most optimizedweather forecasting systems, developed by the European Centre for Medium-Range Weather Forecasts (ECMWF). IFS is built using spectral horizontal discretization and semi-implicit time integration. Whileit is known that spectral models are reaching the limits of scalability, each year new developments come to push the line further, making the model competitive for the years to come. Aiming for further improvements, we note that the use of spherical harmonics is a good fit for the development of exponential time integration techniques. Alongside, exponential integration is an important part of parallel-in-time developments. Concurrently, it is also of particular interest to see the behaviour of a semi-Lagrangian exponential scheme in a more realistic full dynamical core, and this development is an important step to bring the recently developed methods closer to production in weather forecasting. Overall, we expect the project to bring insights into potential novel algorithmic computational gains exploring the time discretization methods, focused on time parallelism.

News published in Agência FAPESP Newsletter about the scholarship:
Articles published in other media outlets (0 total):
More itemsLess items
VEICULO: TITULO (DATA)
VEICULO: TITULO (DATA)

Please report errors in scientific publications list using this form.