Issues in the parallelization of the OORT remeshing toolkit using MPI-based communications
暂无分享,去创建一个
This paper presents the current strategy used in IP-OORT, an ongoing project to extend the application domain of a C++ toolkit library for iterative mesh adaptation. OORT is a class library for sequential structured, unstructured and hybrid mesh adaptation used mainly in the context of CFD computations, that performs iterative mesh refinement, coarsening and smoothing in three dimensions. Extensions to parallelize mesh adaptation for domain decomposition and MPI high-level communication schemes are currently being investigated. This paper presents specific problems encountered in the parallelization of the mesh adaption process such as performance, load balancing and mesh conformity, and the solutions that have been developed in IP-OORT. Some preliminary results are also presented.
[1] Marie-Gabrielle Vallet,et al. Progress on Vertex Relocation Schemes for Structured Grids in a Metric Space , 2002 .
[2] Carl Kesselman,et al. Generalized communicators in the Message Passing Interface , 1996, Proceedings. Second MPI Developer's Conference.
[3] François Guibault,et al. Applying Parmetis to Structured Remeshing for Industrial CFD Applications , 2003, Int. J. High Perform. Comput. Appl..
[4] Jack Dongarra,et al. MPI: The Complete Reference , 1996 .