A mixed precision LOBPCG algorithm

DOI10.1007/S11075-023-01550-9MaRDI QIDQ6141529zbMATH OpenFDO

Authors Daniel Kressner, Yuxin Ma, Meiyue Shao

Publication date 19 December 2023

Published in Numerical Algorithms (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2302.12528

zbMATH Keywords

symmetric eigenvalue problemLOBPCG algorithmmixed precision algorithm

Mathematics Subject Classification ID

Computational methods for sparse matrices (65F50) Numerical computation of eigenvalues and eigenvectors of matrices (65F15) Preconditioners for iterative methods (65F08) Numerical analysis (65-XX)

Abstract: The locally optimal block preconditioned conjugate gradient (LOBPCG) algorithm is a popular approach for computing a few smallest eigenvalues and the corresponding eigenvectors of a large Hermitian positive definite matrix A. In this work, we propose a mixed precision variant of LOBPCG that uses a (sparse) Cholesky factorization of A computed in reduced precision as the preconditioner. To further enhance performance, a mixed precision orthogonalization strategy is proposed. To analyze the impact of reducing precision in the preconditioner on performance, we carry out a rounding error and convergence analysis of PINVIT, a simplified variant of LOBPCG. Our theoretical results predict and our numerical experiments confirm that the impact on convergence remains marginal. In practice, our mixed precision LOBPCG algorithm typically reduces the computation time by a factor of 1.4--2.0 on both CPUs and GPUs.

Recommendations

• Mixed precision s‐step Lanczos and conjugate gradient algorithms
• Mixed precision algorithms in numerical linear algebra
• A robust and efficient implementation of LOBPCG
• Toward the optimal preconditioned eigensolver: Locally optimal block preconditioned conjugate gradient method
• Mixed-precision conjugate gradient algorithm using the groupwise update strategy

Cites work

• scientific article; zbMATH DE number 6159604 (Why is no real title available?)
• A geometric theory for preconditioned inverse iteration applied to a subspace
• A robust and efficient implementation of LOBPCG
• Accelerating the solution of linear systems by iterative refinement in three precisions
• Accuracy and Stability of Numerical Algorithms
• Algorithm 589: SICEDR  : A FORTRAN Subroutine for Improving the Accuracy of Computed Matrix Eigenvalues
• Basis selection in LOBPCG
• Convergence theory for preconditioned eigenvalue solvers in a nutshell
• Iterative refinement for singular value decomposition based on matrix multiplication
• Iterative refinement for symmetric eigenvalue decomposition
• Iterative refinement for symmetric eigenvalue decomposition. II. Clustered eigenvalues
• Iterative refinement of Schur decompositions
• Mixed precision algorithms in numerical linear algebra
• Mixed-Precision Cholesky QR Factorization and Its Case Studies on Multicore CPU with Multiple GPUs
• Numerical methods for large eigenvalue problems
• Perturbed preconditioned inverse iteration for operator eigenvalue problems with applications to adaptive wavelet discretization
• Toward the optimal preconditioned eigensolver: Locally optimal block preconditioned conjugate gradient method

Cited in

(2)

• Mixed-precision conjugate gradient algorithm using the groupwise update strategy
• Mixed Precision Rayleigh Quotient Iteration for Total Least Squares Problems

This page was built for publication: A mixed precision LOBPCG algorithm

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6141529)