Mixed Precision Iterative Refinement with Sparse Approximate Inverse Preconditioning
From MaRDI portal
Publication:6098511
Abstract: With the commercial availability of mixed precision hardware, mixed precision GMRES-based iterative refinement schemes have emerged as popular approaches for solving sparse linear systems. Existing analyses of these approaches, however, are based on using full LU factorizations to construct preconditioners for use within GMRES in each refinement step. In practical applications, inexact preconditioning techniques, such as incomplete LU or sparse approximate inverses, are often used for performance reasons. In this work, we investigate the use of sparse approximate inverse preconditioners based on Frobenius norm minimization within GMRES-based iterative refinement. We analyze the computation of sparse approximate inverses in finite precision and derive constraints under which user-specified stopping criteria will be satisfied. We then analyze the behavior of and convergence constraints for a five-precision GMRES-based iterative refinement scheme that uses sparse approximate inverse preconditioning, which we call SPAI-GMRES-IR. Our numerical experiments confirm the theoretical analysis and illustrate the resulting tradeoffs between preconditioner sparsity and GMRES-IR convergence rate.
Recommendations
- Mixed precision incomplete and factorized sparse approximate inverse preconditioning on GPUs
- A new analysis of iterative refinement and its application to accurate solution of ill-conditioned sparse linear systems
- Accelerating the solution of linear systems by iterative refinement in three precisions
- A fast and robust mixed-precision solver for the solution of sparse symmetric linear systems
- scientific article; zbMATH DE number 6499878
Cites work
- scientific article; zbMATH DE number 1024452 (Why is no real title available?)
- A New Preconditioner that Exploits Low-Rank Approximations to Factorization Error
- A Sparse Approximate Inverse Preconditioner for Nonsymmetric Linear Systems
- A comparative study of sparse approximate inverse preconditioners
- A new analysis of iterative refinement and its application to accurate solution of ill-conditioned sparse linear systems
- Accelerating the solution of linear systems by iterative refinement in three precisions
- Accuracy and Stability of Numerical Algorithms
- Any Nonincreasing Convergence Curve is Possible for GMRES
- Approximate inverse preconditionings for sparse linear systems
- Communication-optimal parallel and sequential QR and LU factorizations
- Factorized Sparse Approximate Inverse Preconditionings I. Theory
- Factorized sparse approximate inverses for preconditioning
- Factorized-sparse-approximate-inverse preconditionings of linear systems with unsymmetric matrices
- Frobenius norm minimization and probing for preconditioning
- GPU-accelerated preconditioned GMRES method for two-dimensional Maxwell's equations
- Iterative Refinement Implies Numerical Stability for Gaussian Elimination
- Iterative Refinement in Floating Point
- Iterative refinement implies numerical stability
- Linear algebra software for large-scale accelerated multicore computing
- Mixed precision algorithms in numerical linear algebra
- Multistage mixed precision iterative refinement
- Orderings for Factorized Sparse Approximate Inverse Preconditioners
- Parallel Preconditioning with Sparse Approximate Inverses
- Preconditioning techniques for large linear systems: A survey
- Solving block low-rank linear systems by LU factorization is numerically stable
- Sparse Approximate-Inverse Preconditioners Using Norm-Minimization Techniques
- Squeezing a Matrix into Half Precision, with an Application to Solving Linear Systems
- The University of Florida sparse matrix collection
- Three-precision GMRES-based iterative refinement for least squares problems
- Using FGMRES to obtain backward stability in mixed precision
Cited in
(14)- Mixed precision GMRES-based iterative refinement with recycling.
- Accelerating the solution of linear systems by iterative refinement in three precisions
- The stability of split-preconditioned FGMRES in four precisions
- Multilevel preconditioning and adaptive sparse solution of inverse problems
- Using FGMRES to obtain backward stability in mixed precision
- Multiresolution approximate inverse preconditioners
- Mixed precision incomplete and factorized sparse approximate inverse preconditioning on GPUs
- Mixed-precision preconditionders in parallel domain decomposition solvers
- Using Mixed Precision for Sparse Matrix Computations to Enhance the Performance while Achieving 64-bit Accuracy
- Avoiding breakdown in incomplete factorizations in low precision arithmetic
- A fast and robust mixed-precision solver for the solution of sparse symmetric linear systems
- Convergence Analysis of a Mixed Precision Parareal Algorithm
- Adaptive Precision Sparse Matrix–Vector Product and Its Application to Krylov Solvers
- Approximate Generalized Inverses with Iterative Refinement for $\epsilon$-Accurate Preconditioning of Singular Systems
This page was built for publication: Mixed Precision Iterative Refinement with Sparse Approximate Inverse Preconditioning
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6098511)
