Seeding and adjoining zero-halo partitioned parallel scientific codes

DOI10.1080/10556788.2019.1591404zbMath1446.90002OpenAlexW2936347648MaRDI QIDQ5113718

Pavanakumar Mohanamuraly, Jens-Dominik Müller, Laurent Hascoët

Publication date: 16 June 2020

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

Full work available at URL: https://qmro.qmul.ac.uk/xmlui/handle/123456789/56467

zbMATH Keywords

fixed-point iterationmessage passingcomputational fluid dynamics (CFD)algorithmic differentiationgradient-based optimizationadjoint source transformationparallel reverse mode

Mathematics Subject Classification ID

Methods of reduced gradient type (90C52) Software, source code, etc. for problems pertaining to operations research and mathematical programming (90-04)

Uses Software

• ADOL-C
• OpenFOAM
• TAPENADE
• TAF
• OP2
• CoDiPack
• SU2
• STAMPS
• AMPI

Cites Work

• High performance parallel computing of flows in complex geometries
• Stabilisation of discrete steady adjoint solvers
• Approximate Riemann solvers, parameter vectors, and difference schemes
• Towards the ultimative conservative difference scheme. V: A second-order sequel to Godunov's method
• The Tapenade automatic differentiation tool
• Evaluating Derivatives
• Recipes for adjoint code construction
• A Fast and High Quality Multilevel Scheme for Partitioning Irregular Graphs
• Automatic Generation of Efficient Adjoint Code for a Parallel Navier-Stokes Solver
• High-Performance Derivative Computations using CoDiPack