A Framework for Error-Bounded Approximate Computing, with an Application to Dot Products

DOI10.1137/21M1406994MaRDI QIDQ5864088zbMATH OpenFDO

Authors James D. Diffenderfer, D. Osei-Kuffuor, Harshitha Menon

Publication date 3 June 2022

Published in SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1137/21M1406994

zbMATH Keywords

approximate computingdot producterror boundhigh-performance computing

Mathematics Subject Classification ID

Numerical linear algebra (65F99) Algorithms with automatic result verification (65G20) Roundoff error (65G50) Numerical algorithms for computer arithmetic, etc. (65Y04) Numerical algorithms for specific classes of architectures (65Y10)

Recommendations

• scientific article; zbMATH DE number 3326064
• Approximate computation of scalar products
• Probabilistic Error Analysis of Approximate Recursive Multipliers
• Efficient and accurate computation of upper bounds of approximation errors
• A note on error bounds for approximation in inner product spaces
• A computable error bound for matrix functionals
• Improved error bounds for inner products in floating-point arithmetic
• A reduced product of absolute and relative error bounds for floating-point analysis
• Improved error bounds for floating-point products and Horner's scheme
• Publication:5751854

Cites work

• scientific article; zbMATH DE number 192975 (Why is no real title available?)
• scientific article; zbMATH DE number 1012640 (Why is no real title available?)
• scientific article; zbMATH DE number 839542 (Why is no real title available?)
• scientific article; zbMATH DE number 6159604 (Why is no real title available?)
• Inexact Newton Methods
• Methods of conjugate gradients for solving linear systems
• Mixed-precision iterative refinement using tensor cores on GPUs to accelerate solution of linear systems
• Rigorous floating-point mixed-precision tuning

Cited in

(4)

• The complexity of error metrics
• Mixed precision algorithms in numerical linear algebra
• Approximate computation of scalar products
• Adaptive Precision Sparse Matrix–Vector Product and Its Application to Krylov Solvers

Describes a project that uses

Uses Software

• zfp
• CUDA
• EnerJ
• CUBLAS
• Thrust
• FPTuner
• GitHub

This page was built for publication: A Framework for Error-Bounded Approximate Computing, with an Application to Dot Products

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5864088)