%0 Conference Paper
%A Lintermann, Andreas
%T EFFICIENT PARALLEL GEOMETRY DISTRIBUTION FOR THE SIMULATION OF COMPLEX FLOWS
%I Institute of Structural Analysis and Antiseismic Research School of Civil Engineering National Technical University of Athens (NTUA) Greece             Athens
%M FZJ-2019-05809
%P 1277-1293
%D 2016
%< Proceedings of the VII European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS Congress 2016) - Institute of Structural Analysis and Antiseismic Research School of Civil Engineering National Technical University of Athens (NTUA) Greece             Athens, 2016. - ISBN 978-618-82844-0-1 - doi:10.7712/100016.1885.5067
%X Highly resolved intrinsic geometrical shapes used in three-dimensional parallel simulations offluid flows consume a large portion ofthe available memory when loaded serially on every process. This demands for a memory efficient implementation of a distributed geometry which is however a non-trivial task when complex spatial domain decomposition methods for the flow domain are involved. To overcome this problem, an algorithm to generate a parallel geometry during the mesh generation is proposed that enables a low-memory subdivision ofthe geometry based on the decomposition of the flow field. The applied meshing method generates computational grids that can be used for simulations on a quasi-arbitrary number ofcores on which the geometry is distributed in an efficient preprocessing step. This allows reducing the number ofinstances ofthe geometry in the global memory ofthesimulation toaboutone. The algorithm is used to generate a parallel geometry for a large shape consisting of 7x10^6 triangles, i.e., for a geometry representing the whole respiratory tract down to the 12th lung generation. For this case, performance and memory consumption measurements are given for simulations on 8,192 up to 131,072 cores and juxtaposed to results obtained from simulations using non-parallel geometries. The findings show that with thenewmethodnot onlythe memory usage could be reduced by the factors of 1,802 and 19,936 for core numbers of 8,192 and 131,072 but also a large speed-up factor ofabout 51 is obtained in the geometry I/O and preprocessing. Furthermore, the parallel geometry allows using the sweet spot with respect to acombination of distributed and sharedmemory parallelization leading to an increase of the computational speed ofabout 1.43.
%B VII European Congress on Computational Methods in Applied Sciences and Engineering
%C 5 Jun 2016 - 10 Jun 2016, Crete Island (Greece)
Y2 5 Jun 2016 - 10 Jun 2016
M2 Crete Island, Greece
%F PUB:(DE-HGF)8 ; PUB:(DE-HGF)7
%9 Contribution to a conference proceedingsContribution to a book
%R 10.7712/100016.1885.5067
%U https://juser.fz-juelich.de/record/866740