A Parallel Direct Eigensolver for Sequences of Hermitian Eigenvalue Problems with No Tridiagonalization
Abstract
In this paper, a Parallel Direct Eigensolver for Sequences of Hermitian Eigenvalue Problems with no tridiagonalization is proposed, denoted by PDESHEP, and it combines direct methods with iterative methods. PDESHEP first reduces a Hermitian matrix to its banded form, then applies a spectrum slicing algorithm to the banded matrix, and finally computes the eigenvectors of the original matrix via backtransform. Therefore, compared with conventional direct eigensolvers, PDESHEP avoids tridiagonalization, which consists of many memory-bounded operations. In this work, the iterative method in PDESHEP is based on the contour integral method implemented in FEAST. The combination of direct methods with iterative methods for banded matrices requires some efficient data redistribution algorithms both from 2D to 1D and from 1D to 2D data structures. Hence, some two-step data redistribution algorithms are proposed, which can be 10× faster than ScaLAPACK routine PXGEMR2D. For the symmetric self-consistent field (SCF) eigenvalue problems, PDESHEP can be on average 1.25× faster than the state-of-the-art direct solver in ELPA when using 4096 processes. Numerical results are obtained for dense Hermitian matrices from real applications and large real sparse matrices from the SuiteSparse collection.