The eigenvalues slicing library (EVSL): algorithms, implementation, and software
From MaRDI portal
Publication:5230632
DOI10.1137/18M1170935zbMATH Open1420.65050arXiv1802.05215OpenAlexW2967701142WikidataQ127375191 ScholiaQ127375191MaRDI QIDQ5230632FDOQ5230632
Authors: Ruipeng Li, Yuanzhe Xi, Lucas Erlandson, Y. Saad 
Publication date: 28 August 2019
Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)
Abstract: This paper describes a software package called EVSL (for EigenValues Slicing Library) for solving large sparse real symmetric standard and generalized eigenvalue problems. As its name indicates, the package exploits spectrum slicing, a strategy that consists of dividing the spectrum into a number of subintervals and extracting eigenpairs from each subinterval independently. In order to enable such a strategy, the methods implemented in EVSL rely on a quick calculation of the spectral density of a given matrix, or a matrix pair. What distinguishes EVSL from other currently available packages is that EVSL relies entirely on filtering techniques. Polynomial and rational filtering are both implemented and are coupled with Krylov subspace methods and the subspace iteration algorithm. On the implementation side, the package offers interfaces for various scenarios including matrix-free modes, whereby the user can supply his/her own functions to perform matrix-vector operations or to solve sparse linear systems. The paper describes the algorithms in EVSL, provides details on their implementations, and discusses performance issues for the various methods.
Full work available at URL: https://arxiv.org/abs/1802.05215
Recommendations
- SLEPc
- A thick-restart Lanczos algorithm with polynomial filtering for Hermitian eigenvalue problems
- Strategies for spectrum slicing based on restarted Lanczos methods
- Computing the eigenvalues of symmetric \(\mathcal{H}^2\)-matrices by slicing the spectrum
- scientific article; zbMATH DE number 1953314
Krylov subspace methodsspectral densityparallel computingpolynomial filteringspectrum slicingrational filtering
Computational methods for sparse matrices (65F50) Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Orthogonalization in numerical linear algebra (65F25)
Cites Work
- LAPACK Users' Guide
- Algorithm 827
- PRIMME: preconditioned iterative multimethod eigensolver -- methods and software description
- Anasazi software for the numerical solution of large-scale eigenvalue problems
- The University of Florida sparse matrix collection
- ARPACK Users' Guide
- PARDISO: a high-performance serial and parallel sparse linear solver in semiconductor device simulation
- Algorithm 832
- JADAMILU: a software code for computing selected eigenvalues of large sparse symmetric matrices
- CIRR: a Rayleigh-Ritz method with contour integral for generalized eigenvalue problems
- Parallel implementations of the trace minimization scheme trace{min} for the sparse symmetric eigenvalue problem
- Cucheb: a GPU implementation of the filtered Lanczos procedure
- Toward the optimal preconditioned eigensolver: Locally optimal block preconditioned conjugate gradient method
- Direct Methods for Sparse Linear Systems
- Chebyshev-filtered subspace iteration method free of sparse diagonalization for solving the Kohn-Sham equation
- Reorthogonalization and Stable Algorithms for Updating the Gram-Schmidt QR Factorization
- Numerical methods for large eigenvalue problems
- A filtered Lanczos procedure for extreme and interior eigenvalue problems
- Title not available (Why is that?)
- An extended set of FORTRAN basic linear algebra subprograms
- Basic Linear Algebra Subprograms for Fortran Usage
- A set of level 3 basic linear algebra subprograms
- A projection method for generalized eigenvalue problems using numerical integration.
- A numerical method for nonlinear eigenvalue problems using contour integrals
- A Trace Minimization Algorithm for the Generalized Eigenvalue Problem
- Rounding error analysis of the classical Gram-Schmidt orthogonalization process
- Thick-restart Lanczos method for large symmetric eigenvalue problems
- SLEPc
- Jacobi--Davidson Style QR and QZ Algorithms for the Reduction of Matrix Pencils
- Self-consistent-field calculations using Chebyshev-filtered subspace iteration
- The loss of orthogonality in the Gram-Schmidt orthogonalization process
- The trace minimization method for the symmetric generalized eigenvalue problem
- Thick-restart Lanczos method for electronic structure calculations
- A spectrum slicing method for the Kohn-Sham problem
- Block Locally Optimal Preconditioned Eigenvalue Xolvers (BLOPEX) in Hypre and PETSc
- State-of-the-art eigensolvers for electronic structure calculations of large scale nano-systems
- Fast computation of spectral densities for generalized eigenvalue problems
- Dynamic Thick Restarting of the Davidson, and the Implicitly Restarted Arnoldi Methods
- Adaptive projection subspace dimension for the thick-restart Lanczos method
- Approximating spectral densities of large matrices
- A thick-restart Lanczos algorithm with polynomial filtering for Hermitian eigenvalue problems
- Computing partial spectra with least-squares rational filters
Cited In (18)
- ELSI -- an open infrastructure for electronic structure solvers
- Proxy-GMRES: Preconditioning via GMRES in Polynomial Space
- Strategies for spectrum slicing based on restarted Lanczos methods
- Hybrid eigensolvers for nuclear configuration interaction calculations
- A new class of AMG interpolation methods based on matrix-matrix multiplications
- Improvements to SLEPc in releases 3.14--3.18
- Rational Spectral Filters with Optimal Convergence Rate
- A non-perturbative approach to computing seismic normal modes in rotating planets
- Limited‐memory polynomial methods for large‐scale matrix functions
- Fast Randomized Non-Hermitian Eigensolvers Based on Rational Filtering and Matrix Partitioning
- On relaxed filtered Krylov subspace method for non-symmetric eigenvalue problems
- EVSL
- A parallel algorithm for computing partial spectral factorizations of matrix pencils via Chebyshev approximation
- A rational filtering algorithm for sequences of shifted symmetric linear systems with applications to frequency response analysis
- An SVD-based fast algorithm for 3D Maxwell's equations with perfect electric conductor and quasi-periodic boundary conditions
- POWEV: A subroutine package to evaluate eigenvalues and eigenvectors of large sparse matrices
- Computational materials science and engineering
- A novel algebraic multigrid approach based on adaptive smoothing and prolongation for ill-conditioned systems
Uses Software
This page was built for publication: The eigenvalues slicing library (EVSL): algorithms, implementation, and software
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5230632)
