High-performance solvers for dense Hermitian eigenproblems
From MaRDI portal
Abstract: We introduce a new collection of solvers - subsequently called EleMRRR - for large-scale dense Hermitian eigenproblems. EleMRRR solves various types of problems: generalized, standard, and tridiagonal eigenproblems. Among these, the last is of particular importance as it is a solver on its own right, as well as the computational kernel for the first two; we present a fast and scalable tridiagonal solver based on the Algorithm of Multiple Relatively Robust Representations - referred to as PMRRR. Like the other EleMRRR solvers, PMRRR is part of the freely available Elemental library, and is designed to fully support both message-passing (MPI) and multithreading parallelism (SMP). As a result, the solvers can equally be used in pure MPI or in hybrid MPI-SMP fashion. We conducted a thorough performance study of EleMRRR and ScaLAPACK's solvers on two supercomputers. Such a study, performed with up to 8,192 cores, provides precise guidelines to assemble the fastest solver within the ScaLAPACK framework; it also indicates that EleMRRR outperforms even the fastest solvers built from ScaLAPACK's components.
Recommendations
- A Parallel Eigensolver for Dense Symmetric Matrices Based on Multiple Relatively Robust Representations
- Improved accuracy and parallelism for MRRR-based eigensolvers -- a mixed precision approach
- Efficient solution of symmetric/Hermitian eigenvalue problems on parallel arrays of vector processors
- Toward an Efficient Parallel Eigensolver for Dense Symmetric Matrices
- ScaLAPACK's MRRR algorithm
Cited in
(12)- High-performance implementation of Chebyshev filter diagonalization for interior eigenvalue computations
- Look-ahead in the two-sided reduction to compact band forms for symmetric eigenvalue problems and the SVD
- EleMRRR
- PMRRR
- Projective block Lanczos algorithm for dense, Hermitian eigensystems
- Improved accuracy and parallelism for MRRR-based eigensolvers -- a mixed precision approach
- scientific article; zbMATH DE number 5503706 (Why is no real title available?)
- Solving dense generalized eigenproblems on multi-threaded architectures
- PHIST: a pipelined, hybrid-parallel iterative solver toolkit
- ChASE: Chebyshev accelerated subspace iteration eigensolver for sequences of Hermitian eigenvalue problems
- Fast computation of spectral projectors of banded matrices
- GPU-acceleration of the ELPA2 distributed eigensolver for dense symmetric and Hermitian eigenproblems
This page was built for publication: High-performance solvers for dense Hermitian eigenproblems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4917162)
