Parallelizable adjoint stencil computations using transposed forward-mode algorithmic differentiation

DOI10.1080/10556788.2018.1435654zbMath1401.65163OpenAlexW2889720240WikidataQ129266547 ScholiaQ129266547MaRDI QIDQ4685562

Jens-Dominik Müller, Jan Hückelheim, Michelle Mills Strout, Paul D. Hovland

Publication date: 9 October 2018

Published in: Optimization Methods and Software (Search for Journal in Brave)

Full work available at URL: http://qmro.qmul.ac.uk/xmlui/handle/123456789/36550

zbMATH Keywords

OpenMPalgorithmic differentiationdiscrete adjointsreverse modeshared-memory parallelism

Mathematics Subject Classification ID

Theory of compilers and interpreters (68N20) Parallel numerical computation (65Y05)

Related Items (1)

Source-to-Source Automatic Differentiation of OpenMP Parallel Loops

Uses Software

• ADIFOR
• fvGCM
• PATUS
• Halide
• Pochoir
• AMPI

Cites Work

• Unnamed Item
• Pseudo-timestepping and verification for automatic differentiation derived CFD codes
• Handling tens of thousands of cores with industrial/legacy codes: approaches, implementation and timings
• Algorithmic Differentiation of Code with Multiple Context-Specific Activities
• On the performance of discrete adjoint CFD codes using automatic differentiation
• Algorithmic Differentiation of Pragma-Defined Parallel Regions

This page was built for publication: Parallelizable adjoint stencil computations using transposed forward-mode algorithmic differentiation