Multi-scale S-fraction reduced-order models for massive wavefield simulations

TL;DR

This paper introduces a novel multi-scale reduced-order modeling approach for large-scale wavefield simulations, leveraging matrix S-fractions to efficiently perform time-domain computations suitable for parallel high-performance computing.

Contribution

The paper presents a new multi-scale ROM framework using matrix S-fractions for wavefield simulations, enabling efficient offline-online computation separation and parallelization.

Findings

Effective reduction in computational complexity for 3D wave simulations

S-fraction based ROMs improve sparsity and parallel efficiency

Numerical results demonstrate promising accuracy and performance

Abstract

We developed a novel reduced-order multi-scale method for solving large time-domain wavefield simulation problems. Our algorithm consists of two main stages. During the first "off-line" stage the fine-grid operator (of the graph Laplacian type} is partitioned on coarse cells (subdomains). Then projection-type multi-scale reduced order models (ROMs) are computed for the coarse cell operators. The off-line stage is embarrassingly parallel as ROM computations for the subdomains are independent of each other. It also does not depend on the number of simulated sources (inputs) and it is performed just once before the entire time-domain simulation. At the second "on-line" stage the time-domain simulation is performed within the obtained multi-scale ROM framework. The crucial feature of our formulation is the representation of the ROMs in terms of matrix Stieltjes continued fractions (S-fractions). The layered structure of the S-fraction introduces several hidden layers in the ROM representation, that results in the block-tridiagonal dynamic system within each coarse cell. This allows us to sparsify the obtained multi-scale subdomain operator ROMs and to reduce the communications between the adjacent subdomains which is highly beneficial for a parallel implementation of the on-line stage. Our approach suits perfectly the high performance computing architectures, however in this paper we present rather promising numerical results for a serial computing implementation only. These results include 3D acoustic and multi-phase anisotropic elastic problems.