A robust and efficient implementation of LOBPCG
From MaRDI portal
Abstract: Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) is widely used to compute eigenvalues of large sparse symmetric matrices. The algorithm can suffer from numerical instability if it is not implemented with care. This is especially problematic when the number of eigenpairs to be computed is relatively large. In this paper we propose an improved basis selection strategy based on earlier work by Hetmaniuk and Lehoucq as well as a robust convergence criterion which is backward stable to enhance the robustness. We also suggest several algorithmic optimizations that improve performance of practical LOBPCG implementations. Numerical examples confirm that our approach consistently and significantly outperforms previous competing approaches in both stability and speed.
Recommendations
- Toward the optimal preconditioned eigensolver: Locally optimal block preconditioned conjugate gradient method
- A refined block conjugate gradient method for symmetric eigenvalue problems
- Basis selection in LOBPCG
- Robust preconditioning of large, sparse, symmetric eigenvalue problems
- Optimize a preconditioned block iterative eigensolver on Sunway Machine
Cites work
- A Block Orthogonalization Procedure with Constant Synchronization Requirements
- An overview of the sparse basic linear algebra subprograms
- Anasazi software for the numerical solution of large-scale eigenvalue problems
- Basic Linear Algebra Subprograms for Fortran Usage
- Basis selection in LOBPCG
- Block Locally Optimal Preconditioned Eigenvalue Xolvers (BLOPEX) in Hypre and PETSc
- Communication lower bounds and optimal algorithms for numerical linear algebra
- Electronic Structure
- scientific article; zbMATH DE number 47363 (Why is no real title available?)
- scientific article; zbMATH DE number 3408799 (Why is no real title available?)
- scientific article; zbMATH DE number 961607 (Why is no real title available?)
- LAPACK Users' Guide
- Roundoff error analysis of the CholeskyQR2 algorithm
- Toward the optimal preconditioned eigensolver: Locally optimal block preconditioned conjugate gradient method
Cited in
(28)- A simple extrapolation method for clustered eigenvalues
- KSSOLV 2.0: an efficient Matlab toolbox for solving the Kohn-Sham equations with plane-wave basis set
- Spectral element methods for eigenvalue problems based on domain decomposition
- Optimize a preconditioned block iterative eigensolver on Sunway Machine
- Randomized Projection for Rank-Revealing Matrix Factorizations and Low-Rank Approximations
- Preconditioned gradient iterations for the eigenproblem of definite matrix pairs
- Fast eigenpairs computation with operator adapted wavelets and hierarchical subspace correction
- Scalable implementation of polynomial filtering for density functional theory calculation in PARSEC
- Hybrid eigensolvers for nuclear configuration interaction calculations
- Hierarchical subspace evolution method for super large parallel computing: A linear solver and an eigensolver as examples
- Householder Orthogonalization with a Nonstandard Inner Product
- A Linearized Structure-Preserving Numerical Scheme for a Gradient Flow Model of the KohnSham Density Functional Theory
- A mixed precision LOBPCG algorithm
- A locally optimal preconditioned Newton-Schur method for symmetric elliptic eigenvalue problems
- Solution of the Schrödinger equation for quasi-one-dimensional materials using helical waves
- On the convergence of orthogonalization-free conjugate gradient method for extreme eigenvalues of Hermitian matrices: a Riemannian optimization interpretation
- A Chebyshev locally optimal block preconditioned conjugate gradient method for product and standard symmetric eigenvalue problems
- High modes smoothing of time schemes for nonlinear parabolic PDEs
- A projection strategy for improving the preconditioner in the LOBPCG
- An improved shifted CholeskyQR based on columns
- A new subspace iteration algorithm for solving generalized eigenvalue problems in vibration analysis
- EPIC: a provable accelerated eigensolver based on preconditioning and implicit convexity
- On a shrink-and-expand technique for symmetric block eigensolvers
- A shifted Laplace rational filter for large-scale eigenvalue problems
- A novel splitting strategy to accelerate solving generalized eigenvalue problem from Kohn-Sham density functional theory
- Subspace embedding with random Khatri-Rao products and its application to eigensolvers
- Jackpot: approximating uncertainty domains with adversarial manifolds
- Basis selection in LOBPCG
This page was built for publication: A robust and efficient implementation of LOBPCG
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4685340)
