Machine Learning Techniques for Data Reduction of CFD Applications

TL;DR

This paper introduces a guaranteed block autoencoder method leveraging tensor correlations for efficient data reduction in CFD applications, achieving high compression ratios while maintaining acceptable error bounds.

Contribution

It presents a novel tensor-based autoencoder approach with error guarantees for CFD data reduction, outperforming existing methods like SZ in compression efficiency.

Findings

Achieves two orders of magnitude data reduction

Maintains errors within scientifically acceptable bounds

Outperforms SZ-based reduction methods

Abstract

We present an approach called guaranteed block autoencoder that leverages Tensor Correlations (GBATC) for reducing the spatiotemporal data generated by computational fluid dynamics (CFD) and other scientific applications. It uses a multidimensional block of tensors (spanning in space and time) for both input and output, capturing the spatiotemporal and interspecies relationship within a tensor. The tensor consists of species that represent different elements in a CFD simulation. To guarantee the error bound of the reconstructed data, principal component analysis (PCA) is applied to the residual between the original and reconstructed data. This yields a basis matrix, which is then used to project the residual of each instance. The resulting coefficients are retained to enable accurate reconstruction. Experimental results demonstrate that our approach can deliver two orders of magnitude in reduction while still keeping the errors of primary data under scientifically acceptable bounds. Compared to reduction-based approaches based on SZ, our method achieves a substantially higher compression ratio for a given error bound or a better error for a given compression ratio.