TY - GEN
T1 - Parallel hierarchical unstructured mesh generation with general cutting
AU - Karman, Steve L.
AU - Betro, Vincent C.
PY - 2008
Y1 - 2008
N2 - A parallel mesh generation process was created to allow for rapid, automatic generation of inviscid meshes for arbitrary, complex geometry. This procedure uses Octree refinement of a Cartesian mesh with general cutting to create hybrid unstructured meshes. The meshing is completed in parallel using OpenMPI as the message passing protocol. Partitioning is performed using the Octree data structure or the METIS partitioning library, depending on the voxel/processor ratio. Multiple cases were run to test the robustness of the algorithm, including surface ships, airplanes, axial compressors and geographic landscapes. Parallel performance is documented in all 3D cases from 1 to 244 processors.
AB - A parallel mesh generation process was created to allow for rapid, automatic generation of inviscid meshes for arbitrary, complex geometry. This procedure uses Octree refinement of a Cartesian mesh with general cutting to create hybrid unstructured meshes. The meshing is completed in parallel using OpenMPI as the message passing protocol. Partitioning is performed using the Octree data structure or the METIS partitioning library, depending on the voxel/processor ratio. Multiple cases were run to test the robustness of the algorithm, including surface ships, airplanes, axial compressors and geographic landscapes. Parallel performance is documented in all 3D cases from 1 to 244 processors.
UR - http://www.scopus.com/inward/record.url?scp=78149437512&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:78149437512
SN - 9781563479373
T3 - 46th AIAA Aerospace Sciences Meeting and Exhibit
BT - 46th AIAA Aerospace Sciences Meeting and Exhibit
T2 - 46th AIAA Aerospace Sciences Meeting and Exhibit
Y2 - 7 January 2008 through 10 January 2008
ER -