Using Mixed Precision for Sparse Matrix Computations to Enhance the Performance while Achieving 64-bit Accuracy

DOI10.1145/1377596.1377597MaRDI QIDQ3549231zbMATH OpenOpenAlexWikidataFDO

Authors Alfredo Buttari, Jakub Kurzak, Piotr Luszczek, Jack Dongarra, Stanimire Tomov

Publication date 21 December 2008

Published in ACM Transactions on Mathematical Software (Search for Journal in Brave)

Full work available at URL http://eprints.maths.manchester.ac.uk/820/1/Using_Mixed_Precision.pdf

Mathematics Subject Classification ID

Direct numerical methods for linear systems and matrix inversion (65F05) Roundoff error (65G50) Iterative numerical methods for linear systems (65F10) Mesh generation, refinement, and adaptive methods for ordinary differential equations (65L50)

Recommendations

• Accelerating scientific computations with mixed precision algorithms
• Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems
• A fast and robust mixed-precision solver for the solution of sparse symmetric linear systems
• Mixed precision algorithms in numerical linear algebra
• Mixed Precision Iterative Refinement with Sparse Approximate Inverse Preconditioning

Cited in

(15)

• Mixed precision bisection
• Optimizing sparse matrix-matrix multiplication for the GPU
• Mixed precision algorithms in numerical linear algebra
• Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems
• Parallel symmetric sparse matrix-vector product on scalar multi-core CPUs
• Avoiding breakdown in incomplete factorizations in low precision arithmetic
• Linear algebra software for large-scale accelerated multicore computing
• Combining sparse approximate factorizations with mixed-precision iterative refinement
• Algebraic Error Analysis for Mixed-Precision Multigrid Solvers
• A hybridizable discontinuous Galerkin method combined to a Schwarz algorithm for the solution of 3d time-harmonic Maxwell's equation
• Parallel algebraic hybrid solvers for large 3D convection-diffusion problems
• Accelerating scientific computations with mixed precision algorithms
• Towards dense linear algebra for hybrid GPU accelerated manycore systems
• Algebraic Error Analysis for Mixed-Precision Multigrid Solvers
• A note on generating finer‐grain parallelism in a representation tree

Describes a project that uses

Uses Software

• mctoolbox
• PETSc
• DQGMRES
• LAPACK
• SuperLU-DIST

This page was built for publication: Using Mixed Precision for Sparse Matrix Computations to Enhance the Performance while Achieving 64-bit Accuracy

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3549231)