Automating non-unimodular loop transformations for massive parallelism

DOI10.1016/0167-8191(94)90002-7MaRDI QIDQ1323638zbMATH OpenOpenAlexFDO

Authors Jingling Xue

Publication date 26 January 1995

Published in Parallel Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/0167-8191(94)90002-7

zbMATH Keywords

latticeHermite normal formloop transformationsFourier-Motzkin methodnon-unimodularparallelising compilationregular array design

Mathematics Subject Classification ID

Parallel numerical computation (65Y05) Modes of computation (nondeterministic, parallel, interactive, probabilistic, etc.) (68Q10) Theory of compilers and interpreters (68N20)

Recommendations

• On loop transformations of nested loops with affine dependencies
• Beyond unimodular transformations
• Unimodular transformations of non-perfectly nested loops
• scientific article; zbMATH DE number 618313

Cited in

(4)

• Automatic synthesis of parallel algorithms
• Semi-automatic composition of loop transformations for deep parallelism and memory hierarchies
• A singular loop transformation framework based on non-singular matrices
• Space-Time Equations for Non-Unimodular Mappings

This page was built for publication: Automating non-unimodular loop transformations for massive parallelism

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1323638)