Second Order Accurate Hierarchical Approximate Factorization of Sparse SPD Matrices

Abstract

We describe a second-order accurate approach to sparsifying the off-diagonal blocks in the hierarchical approximate factorizations of sparse symmetric positive definite matrices. The norm of the error made by the new approach depends quadratically, not linearly, on the error in the low-rank approximation of the given block. The analysis of the resulting two-level preconditioner shows that the preconditioner is second-order accurate as well. We incorporate the new approach into the recent Sparsified Nested Dissection algorithm [SIAM J. Matrix Anal. Appl., 41 (2020), pp. 715-746 ], and test it on a wide range of problems. The new approach halves the number of Conjugate Gradient iterations needed for convergence, with almost the same factorization complexity, improving the total runtimes of the algorithm. Our approach can be incorporated into other rank-structured methods for solving sparse linear systems.

Hierarchical Orthogonal Factorization: Sparse Least Squares Problems

“…Hierarchical solvers such as Hierarchical Interpolative Factorization (HIF) [23,22,11,12], LoRaSp [35,42] and Sparsified Nested Dissection (spaND) [5,29] are another family of incomplete factorizations. All three solvers were developed to perform a fast Cholesky factorization of SPD matrices by incorporating low-rank approximations in the classical multifrontal approach.…”