A new GPU implementation for lattice-Boltzmann simulations on sparse geometries

11/08/2016
by   Tadeusz Tomczak, et al.
0

We describe a high-performance implementation of the lattice Boltzmann method (LBM) for sparse 3D geometries on graphic processors (GPU). The main contribution of this work is a data layout that allows to minimise the number of redundant memory transactions during the propagation step of LBM. We show that by using a uniform mesh of small three-dimensional tiles and a careful data placement it is possible to utilise more than 70 GPU memory bandwidth for D3Q19 lattice and double precision numbers. The performance of our implementation is thoroughly examined and compared with other GPU implementations of LBM. The proposed method performs the best for sparse geometries with good spatial locality.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
08/30/2021

Data-Oriented Language Implementation of Lattice-Boltzmann Method for Dense and Sparse Geometries

The performance of lattice-Boltzmann solver implementations usually depe...
research
04/13/2023

Thread-safe lattice Boltzmann for high-performance computing on GPUs

We present thread-safe, highly-optimized lattice Boltzmann implementatio...
research
10/22/2021

GPU-Accelerated Forward-Backward algorithm with Application to Lattice-Free MMI

We propose to express the forward-backward algorithm in terms of operati...
research
08/07/2023

GPU Optimization of Lattice Boltzmann Method with Local Ensemble Transform Kalman Filter

The ensemble data assimilation of computational fluid dynamics simulatio...
research
03/27/2019

Efficient LBM on GPUs for dense moving objects using immersed boundary condition

There exists an increasing interest for using immersed boundary methods ...
research
10/22/2020

Cross-platform programming model for many-core lattice Boltzmann simulations

We present a novel, hardware-agnostic implementation strategy for lattic...

Please sign up or login with your details

Forgot password? Click here to reset