GPU acceleration of average gradient method for solving partial differential equations

Touko Puro, Aarne Pohjonen

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingsScientificpeer-review

5 Downloads (Pure)

Abstract

Previously presented method of calculating local average gradients for solvingpartial differential equations (PDEs) is enhanced by accelerating it with graphics processingunits (GPUs) and combining a previous technique of interpolating between grid points in thecalculation of the gradients instead of using interpolation to create a denser grid.For accelerating the calculation with GPUs, we have ported the original naive Matlabimplementation to C++ and CUDA, and after optimizing the code we observe a speedupfactors more than two thousand, which is largely due to the original code not being optimized.
Original languageEnglish
Title of host publicationProceedings of SIMS EUROSIM 2024 Oulu, Finland, 11-12 September, 2024
PublisherLinköping University Electronic Press
Pages488-494
ISBN (Electronic)978-91-8075-984-7
DOIs
Publication statusPublished - 13 Jan 2025
MoE publication typeA4 Conference publication
Event2nd SIMS EUROSIM Conference on Simulation and Modelling and 65th SIMS Conference on Simulation and Modelling - Oulu, Finland
Duration: 11 Sept 202412 Sept 2024

Publication series

NameLinköping Electronic Conference Proceedings
PublisherLinköping University Electronic Press
Volume212
ISSN (Print)1650-3686
ISSN (Electronic)1650-3740

Conference

Conference2nd SIMS EUROSIM Conference on Simulation and Modelling and 65th SIMS Conference on Simulation and Modelling
Country/TerritoryFinland
CityOulu
Period11/09/202412/09/2024

Fingerprint

Dive into the research topics of 'GPU acceleration of average gradient method for solving partial differential equations'. Together they form a unique fingerprint.

Cite this