CSDMS 2023: Patterns and Processes Across Scales

Using GPUs to Solve Science Problems Faster

Richard Barnes

Lawrence Berkeley National Lab, United States

GPUs can make models, simulations, machine learning, and data analysis much faster, but how? And when? In this clinic we'll discuss whether you should use a GPU for your work, whether you should buy one, which one to buy, and how to use one effectively. We'll also get hands-on and speed up a landscape evolution model together. This clinic should be of interest both to folks who would like to speed up their code with minimal effort as well as folks who are interested in the nitty gritty of pushing computational boundaries.

Please acknowledge the original contributors when you are using this material. If there are any copyright issues, please let us know ( and we will respond as soon as possible.

Of interest for:
  • Marine Working Group
  • Terrestrial Working Group
  • Coastal Working Group
  • Cyberinformatics and Numerics Working Group
  • Hydrology Focus Research Group
  • Geodynamics Focus Research Group
  • Artificial Intelligence & Machine Learning Initiative
  • Modeling Platform Interoperability Initiative
  • River Network Modeling Initiative