English

$O(N)$ distributed direct factorization of structured dense matrices using runtime systems

Numerical Analysis 2023-11-03 v1 Mathematical Software Numerical Analysis

Abstract

Structured dense matrices result from boundary integral problems in electrostatics and geostatistics, and also Schur complements in sparse preconditioners such as multi-frontal methods. Exploiting the structure of such matrices can reduce the time for dense direct factorization from O(N3)O(N^3) to O(N)O(N). The Hierarchically Semi-Separable (HSS) matrix is one such low rank matrix format that can be factorized using a Cholesky-like algorithm called ULV factorization. The HSS-ULV algorithm is highly parallel because it removes the dependency on trailing sub-matrices at each HSS level. However, a key merge step that links two successive HSS levels remains a challenge for efficient parallelization. In this paper, we use an asynchronous runtime system PaRSEC with the HSS-ULV algorithm. We compare our work with STRUMPACK and LORAPO, both state-of-the-art implementations of dense direct low rank factorization, and achieve up to 2x better factorization time for matrices arising from a diverse set of applications on up to 128 nodes of Fugaku for similar or better accuracy for all the problems that we survey.

Keywords

Cite

@article{arxiv.2311.00921,
  title  = {$O(N)$ distributed direct factorization of structured dense matrices using runtime systems},
  author = {Sameer Deshmukh and Qinxiang Ma and Rio Yokota and George Bosilca},
  journal= {arXiv preprint arXiv:2311.00921},
  year   = {2023}
}
R2 v1 2026-06-28T13:09:11.315Z