Mixed Precision Block Fused Multiply-Add: Error Analysis and Application to GPU Tensor Cores
From MaRDI portal
Publication:3300847
DOI10.1137/19M1289546zbMath1452.65425MaRDI QIDQ3300847
Pierre Blanchard, Nicholas J. Higham, Theo A. Mary, Florent Lopez, Srikara Pranesh
Publication date: 30 July 2020
Published in: SIAM Journal on Scientific Computing (Search for Journal in Brave)
Roundoff error (65G50) Direct numerical methods for linear systems and matrix inversion (65F05) Numerical algorithms for specific classes of architectures (65Y10) Preconditioners for iterative methods (65F08)
Related Items
Matrix Multiplication in Multiword Arithmetic: Error Analysis and Application to GPU Tensor Cores, Mixed precision algorithms in numerical linear algebra, Mixed-precision explicit stabilized Runge-Kutta methods for single- and multi-scale differential equations, Double precision is not necessary for LSQR for solving discrete linear ill-posed problems, Numerical stability of algorithms at extreme scale and low precisions, Sharper Probabilistic Backward Error Analysis for Basic Linear Algebra Kernels with Random Data, Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems, Numerical algorithms for high-performance computational science, Rounding Error Analysis of Mixed Precision Block Householder QR Algorithms, Exploiting Lower Precision Arithmetic in Solving Symmetric Positive Definite Linear Systems and Least Squares Problems
Uses Software
Cites Work
- Verification methods: Rigorous results using floating-point arithmetic
- IEEE754 Precision- k base-β Arithmetic Inherited by Precision- m Base-β Arithmetic for k < m
- The Arithmetic of the Digital Computer: A New Approach
- Systolic super summation
- 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
- A Class of Fast and Accurate Summation Algorithms
- Squeezing a Matrix into Half Precision, with an Application to Solving Linear Systems
- A New Approach to Probabilistic Rounding Error Analysis
