Compensated summation and dot product algorithms for floating-point vectors on parallel architectures: error bounds, implementation and application in the Krylov subspace methods

DOI10.1016/j.cam.2022.114434zbMath1492.65374OpenAlexW4281636333WikidataQ114201790 ScholiaQ114201790MaRDI QIDQ2151619

N. M. Evstigneev, O. I. Ryabkov, V. P. Petrovskiy, I. O. Teplyakov, A. N. Bocharov

Publication date: 5 July 2022

Published in: Journal of Computational and Applied Mathematics (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.cam.2022.114434

zbMATH Keywords

accurate dot productaccurate summationKrylov subspace solverscompensated algorithmsgeneral purpose GPU

Mathematics Subject Classification ID

Roundoff error (65G50) Parallel numerical computation (65Y05) Numerical algorithms for specific classes of architectures (65Y10) Preconditioners for iterative methods (65F08)

Uses Software

• CUDA
• mctoolbox
• XBLAS
• Compqd

Cites Work

• Unnamed Item
• Accurate quotient-difference algorithm: error analysis, improvements and applications
• Implicit method for the solution of supersonic and hypersonic 3D flow problems with lower-upper symmetric-Gauss-Seidel preconditioner on multiple graphics processing units
• Reproducibility strategies for parallel preconditioned conjugate gradient
• A floating-point technique for extending the available precision
• Residue Number Systems
• Accurate Parallel Floating-Point Accumulation
• The Accuracy of Floating Point Summation
• Accurate Floating-Point Summation Part II: Sign, K-Fold Faithful and Rounding to Nearest
• Floating-Point Computation of Functions with Maximum Accuracy
• A Distillation Algorithm for Floating-Point Summation
• Emulation of a FMA and Correctly Rounded Sums: Proved Algorithms Using Rounding to Odd
• Handbook of Floating-Point Arithmetic
• Accuracy and Stability of Numerical Algorithms
• A Class of Fast and Accurate Summation Algorithms
• Accurate Floating-Point Summation Part I: Faithful Rounding
• Design, implementation and testing of extended and mixed precision BLAS
• On accurate floating-point summation
• Numerical Stability in Problems of Linear Algebra
• Accurate Sum and Dot Product

This page was built for publication: Compensated summation and dot product algorithms for floating-point vectors on parallel architectures: error bounds, implementation and application in the Krylov subspace methods