GBEES-GPU: An efficient parallel GPU algorithm for high-dimensional nonlinear uncertainty propagation

Eulerian nonlinear uncertainty propagation methods often suffer from finite domain limitations and computational inefficiencies. A recent approach to this class of algorithm, Grid-based Bayesian Estimation Exploiting Sparsity, addresses the first challenge by dynamically allocating a discretized gri...

Full description

Saved in:
Bibliographic Details
Published in:Computer physics communications Vol. 317; p. 109819
Main Authors: Hanson, Benjamin L., Rubio, Carlos, García-Gutiérrez, Adrián, Bewley, Thomas
Format: Journal Article
Language:English
Published: Elsevier B.V 01.12.2025
Subjects:
Corner transport upwind
CUDA
Dynamic gridding
Eulerian uncertainty propagation
GPU
Hashtables
CUDA
Dynamic gridding
GPU
Corner transport upwind
Eulerian uncertainty propagation
Hashtables
ISSN:0010-4655
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Eulerian nonlinear uncertainty propagation methods often suffer from finite domain limitations and computational inefficiencies. A recent approach to this class of algorithm, Grid-based Bayesian Estimation Exploiting Sparsity, addresses the first challenge by dynamically allocating a discretized grid in regions of phase space where probability is non-negligible. However, the design of the original algorithm causes the second challenge to persist in high-dimensional systems. This paper presents an architectural optimization of the algorithm for CPU implementation, followed by its adaptation to the CUDA framework for single GPU execution. The algorithm is validated for accuracy and convergence, with performance evaluated across distinct GPUs. Tests include propagating a three-dimensional probability distribution subject to the Lorenz '63 model and a six-dimensional probability distribution subject to the Lorenz '96 model. The results imply that the improvements made result in a speedup of over 1000 times compared to the original implementation. •The Grid-based Bayesian Estimation Exploiting Sparsity (GBEES) algorithm with poor time complexity is optimized for the CPU.•The optimized CPU version (GBEES-CPU-optimized) is then translated to CUDA for GPU execution (GBEES-GPU).•GBEES-CPU-optimized and GBEES-GPU are validated on 3D and 6D chaotic Lorenz systems with a dense MC distribution.•GBEES-CPU-optimized achieves an estimated speedup of one order of magnitude when compared with the legacy implementation.•GBEES-GPU achieves an estimated speedup of three orders of magnitude when compared with the legacy implementation.
ISSN:0010-4655
DOI:10.1016/j.cpc.2025.109819