Reducing the memory usage of Lattice-Boltzmann schemes with a DWT-based compression

TL;DR

This paper introduces a DWT-based lossy compression method to significantly reduce memory consumption in high-performance Lattice-Boltzmann and Finite-Volume simulations, maintaining accuracy while achieving high compression ratios.

Contribution

It presents a novel DWT-based lossy compression scheme specifically designed for Lattice-Boltzmann and Finite-Volume schemes to reduce memory usage without sacrificing accuracy.

Findings

Memory usage reduced by several orders of magnitude.

High compression ratios achieved with preserved simulation accuracy.

Applicable to multiple FV/LBM schemes.

Abstract

This paper presents a new solution to address the challenge of increasing memory usage in high-performance computing simulations of Lattice-Bolzmann or Finite-Volume schemes.Our approach utilises a lossy compression scheme based on the Discrete Wavelet Transform (DWT) to achieve high compression ratios while preserving the accuracy of the simulation.Our evaluation on two different FV/LBM schemes demonstrates that the approach can reduce memory usage by several orders of magnitude.