Uniform (mathcal {H})-Matrix Compression with Applications to Boundary Integral Equations

📅 2024-05-24
🏛️ SIAM Journal on Scientific Computing
📈 Citations: 0
Influential: 0
📄 PDF

career value

218K/year
🤖 AI Summary
Discretizing boundary integral equations yields dense matrices; a key challenge is reducing memory consumption and computational time while preserving $mathcal{O}(Nlog N)$ storage and matrix–vector multiplication complexity. Method: We propose the algebraic Uniform $mathcal{H}$-matrix format—a novel sparse representation framework situated between standard $mathcal{H}$- and $mathcal{H}^2$-matrices—enabling fully algebraic, geometry-free compression via hierarchical clustering, low-rank subblock approximation, and boundary element discretization. The method is applicable to oscillatory kernel problems such as the Helmholtz equation. Contribution/Results: To our knowledge, this is the first purely algebraic $mathcal{H}$-matrix variant achieving uniform complexity without geometric input. Experiments on medium-scale problems show ~30% reduction in matrix storage, 1.8× speedup in matrix–vector multiplication, and negligible overhead in construction time—outperforming state-of-the-art $mathcal{H}$- and $mathcal{H}^2$-matrix methods in practice.

Technology Category

Application Category

📝 Abstract
Boundary integral equations lead to dense system matrices when discretized, yet they are data-sparse. Using the $mathcal{H}$-matrix format, this sparsity is exploited to achieve $mathcal{O}(Nlog N)$ complexity for storage and multiplication by a vector. This is achieved purely algebraically, based on low-rank approximations of subblocks, and hence the format is also applicable to a wider range of problems. The $mathcal{H}^2$-matrix format improves the complexity to $mathcal{O}(N)$ by introducing a recursive structure onto subblocks on multiple levels. However, in many cases this comes with a large proportionality constant, making the $mathcal{H}^2$-matrix format advantageous mostly for large problems. In this paper we investigate the usefulness of a matrix format that lies in between these two: Uniform $mathcal{H}$-matrices. An algebraic compression algorithm is introduced to transform a regular $mathcal{H}$-matrix into a uniform $mathcal{H}$-matrix, which maintains the asymptotic complexity. Using examples of the BEM formulation of the Helmholtz equation, we show that this scheme lowers the storage requirement and execution time of the matrix-vector product without significantly impacting the construction time.
Problem

Research questions and friction points this paper is trying to address.

Compress dense matrices from boundary integral equations efficiently
Reduce storage and computation costs in H-matrix formats
Balance complexity and performance between H and H2 matrices
Innovation

Methods, ideas, or system contributions that make the work stand out.

Uses uniform H-matrices for intermediate complexity
Applies algebraic compression to H-matrices
Reduces storage and execution time efficiently
🔎 Similar Papers