Using coordination to parallelize sparse-grid methods for 3D CFD problems
The good parallel computing properties of sparse-grid solution techniques are investigated. For this, an existing sequential CFD code for a standard 3D problem from computational aerodynamics is restructured into a parallel application. The restructuring is organized according to a master/slave protocol. The coordinator modules developed thereby are implemented in the coordination language Manifold and are generally applicable. Performance results are given for both the sequential and parallel version of the code. The results are promising, the paper contributes to the state-of-the-art in improving the efficiency of large-scale computations. Also a theoretical analysis is made of speed-up through parallelization in a multi-user single-machine environment.
|Concurrent Programming (acm D.1.3), Language Classifications (acm D.3.2), Language Constructs and Features (acm D.3.3), Modes of Computation (acm F.1.2), Languages and Systems (acm I.1.3)|
|Mesh generation and refinement (msc 65N50), Multigrid methods; domain decomposition (msc 65N55), None of the above, but in MSC2010 section 65Nxx (msc 65N99), Other numerical methods (msc 76M25), Gas dynamics, general (msc 76N15)|
|Software (theme 1), Life Sciences (theme 5), Energy (theme 4)|
|Software Engineering [SEN]|
Everaars, C.T.H, & Koren, B. (1997). Using coordination to parallelize sparse-grid methods for 3D CFD problems. Software Engineering [SEN]. CWI.