Accelerated Thomas Solver for (Quasi-)Block-Tridiagonal Linear Algebraic Equation Systems, Using SSE/AVX Instruction Sets for Vectorizing Dense Block Operations

DOI10.1142/S021987621750027XzbMath1359.65037MaRDI QIDQ2972177

Dawid Barnaś, Lesław K. Bieniasz

Publication date: 7 April 2017

Published in: International Journal of Computational Methods (Search for Journal in Brave)

zbMATH Keywords

Thomas algorithmcomputational electrochemistrystreaming SIMD extensionsadvanced vector extensionsquasi-block-tridiagonal

Mathematics Subject Classification ID

Parallel numerical computation (65Y05) Direct numerical methods for linear systems and matrix inversion (65F05)

Uses Software

• SPIKE
• Emmerald

Cites Work

• Erratum to ``Extension of the Thomas algorithm to a class of algebraic linear equation systems involving quasi-block-tridiagonal matrices with isolated block-pentadiagonal rows, assuming variable block dimensions
• High order accurate, one-sided finite-difference approximations to concentration gradients at the boundaries, for the simulation of electrochemical reaction-diffusion problems in one-dimensional space geometry.
• Block tridiagonal matrix inversion and fast transmission calculations
• Emmerald: a fast matrix–matrix multiply using Intel's SSE instructions
• Unnamed Item
• Unnamed Item
• Unnamed Item

This page was built for publication: Accelerated Thomas Solver for (Quasi-)Block-Tridiagonal Linear Algebraic Equation Systems, Using SSE/AVX Instruction Sets for Vectorizing Dense Block Operations