English
 
Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Conference Paper

Introducing Underworld3 - a High Performance Geodynamic Modelling Code Focused on Usability and Reusability

Authors

Moresi,  Louis
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Giordani,  Julian
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

Knight,  Ben
IUGG 2023, General Assemblies, 1 General, International Union of Geodesy and Geophysics (IUGG), External Organizations;

External Ressource
No external resources are shared
Fulltext (public)
There are no public fulltexts stored in GFZpublic
Supplementary Material (public)
There is no public supplementary material available
Citation

Moresi, L., Giordani, J., Knight, B. (2023): Introducing Underworld3 - a High Performance Geodynamic Modelling Code Focused on Usability and Reusability, XXVIII General Assembly of the International Union of Geodesy and Geophysics (IUGG) (Berlin 2023).
https://doi.org/10.57757/IUGG23-3770


Cite as: https://gfzpublic.gfz-potsdam.de/pubman/item/item_5020775
Abstract
We introduce underworld3, a finite element, solid-Earth modelling code designed to be both straightforward to use and highly scalable to peak high-performance computing environments. underworld3 inherits the design patterns of earlier versions of underworld such as 1) A python user interface that is inherently safe for parallel computation. 2) A symbolic interface that allows users to construct and simplify combinations of functions, unknowns and gradients-of-unknowns on the fly that are then used to construct the finite element representation of the problem. 3) Fast, robust, parallel numerical solvers, 4) Flexible, Lagrangian "particle" swarms for modeling transport-dominated quantities that are fully interchangeable with other data-types and can be treated as symbolic quantities. 5) Mesh adaptivity that falls fully within the symbolic framework. The design philosophy of Underworld is to provide readable python code that stays close to the mathematics of the problem definition and is scalable (unedited) from Jupyter notebooks all the way to tens of thousands of cores. Underworld is entirely open source and founded on open source computational frameworks. It provides an accessible entry point for anyone to begin modelling (on a laptop or a free cloud resource) and those models can easily be scaled-up to solve cutting edge problems of societal importance. Typical problems of interest are groundwater flow studies on a whole-of-basin scale, earthquake rupture models coupled with tectonic deformation with applications to hazards, ice-sheet flow and deformation, subduction zone models coupled to over-riding plate deformation, and stress modelling at the continental scale.