Reducing Floating Point Error in Dot Product Using the Superblock Family of Algorithms

DOI10.1137/070679946zbMath1189.65076OpenAlexW2078152929MaRDI QIDQ110775

Anthony M. Castaldo, Anthony T. Chronopoulos, R. Clint Whaley, R. Clint Whaley, Anthony M. Castaldo, Anthony Theodore Chronopoulos

Publication date: January 2009

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

Full work available at URL: https://semanticscholar.org/paper/2483c58592a7651e8f82db540cf19e9ce219f9f2

zbMATH Keywords

algorithmsnumerical exampleserror analysiserror boundsinner productBLASmemory usagecomputational performancedot producterror behaviorATLASerror-reducing algorithmssuperblock

Mathematics Subject Classification ID

Roundoff error (65G50) Complexity and performance of numerical algorithms (65Y20)

Related Items

Minimizing synchronizations in sparse iterative solvers for distributed supercomputers ⋮ Verified numerical computations for large-scale linear systems. ⋮ Floating-point arithmetic ⋮ A Class of Fast and Accurate Summation Algorithms ⋮ Numerical stability of algorithms at extreme scale and low precisions ⋮ Sharper Probabilistic Backward Error Analysis for Basic Linear Algebra Kernels with Random Data ⋮ PreciseSums ⋮ A New Approach to Probabilistic Rounding Error Analysis

Uses Software

• ATLAS
• ScaLAPACK