Automatic exploration of reduced floating-point representations in iterative methods

DOI10.1007/978-3-030-29400-7_34MaRDI QIDQ3297575zbMATH OpenOpenAlexFDO

Authors Yohan Chatelain, Eric Petit, Pablo de Oliveira Castro, David Defour, Ghislain Lartigue

Publication date 20 July 2020

Published in Lecture Notes in Computer Science (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/978-3-030-29400-7_34

Mathematics Subject Classification ID

Iterative numerical methods for linear systems (65F10) Numerical solution of discretized equations for boundary value problems involving PDEs (65N22) Parallel numerical computation (65Y05) Direct numerical and large eddy simulation of turbulence (76F65)

Recommendations

• Rigorous floating-point mixed-precision tuning
• scientific article; zbMATH DE number 2151189
• scientific article; zbMATH DE number 1941105
• Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems
• FloatX

Cites work

• A multi-grid framework for the extraction of large-scale vortices in large-eddy simulation
• Accuracy and Stability of Numerical Algorithms
• Black box multigrid
• Deflation of Conjugate Gradients with Applications to Boundary Value Problems
• Design of a massively parallel CFD code for complex geometries
• Emulation of a FMA and Correctly Rounded Sums: Proved Algorithms Using Rounding to Odd
• Large-eddy simulation of wind turbines wakes including geometrical effects
• Numerical Solution of the Navier-Stokes Equations
• Progress-variable approach for large-eddy simulation of non-premixed turbulent combustion

Cited in

(5)

• Fast and efficient bit-level precision tuning
• scientific article; zbMATH DE number 1941105 (Why is no real title available?)
• scientific article; zbMATH DE number 2151189 (Why is no real title available?)
• Stability Analysis of Inline ZFP Compression for Floating-Point Data in Iterative Methods
• Rigorous floating-point mixed-precision tuning

Describes a project that uses

Uses Software

• mctoolbox
• FloPoCo
• YALES2
• AddressSanitizer

This page was built for publication: Automatic exploration of reduced floating-point representations in iterative methods

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3297575)