A framework for hybrid parallel flow simulations with a trillion cells in complex geometries. Godenschwager, C., Schornbaum, F., Bauer, M., Köstler, H., & Rüde, U. Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '13, IEEE, 1, 2013.
Paper doi abstract bibtex waLBerla is a massively parallel software framework for simulating complex flows with the lattice Boltzmann method (LBM). Performance and scalability results are presented for SuperMUC, the world's fastest x86-based supercomputer ranked number 6 on the Top500 list, and JUQUEEN, a Blue Gene/Q system ranked as number 5. We reach resolutions with more than one trillion cells and perform up to 1.93 trillion cell updates per second using 1.8 million threads. The design and implementation of waLBerla is driven by a careful analysis of the performance on current petascale supercomputers. Our fully distributed data structures and algorithms allow for efficient, massively parallel simulations on these machines. Elaborate node level optimizations and vectorization using SIMD instructions result in highly optimized compute kernels for the single- and two-relaxation-time LBM. Excellent weak and strong scaling is achieved for a complex vascular geometry of the human coronary tree.
@article{Godenschwager-2013-ID254,
title = {A framework for hybrid parallel flow simulations with a trillion cells in
complex geometries},
abstract = {wa{LB}erla is a massively parallel software framework for simulating
complex flows with the lattice Boltzmann method ({LBM}). Performance and
scalability results are presented for Super{MUC}, the world's fastest
x86-based supercomputer ranked number 6 on the Top500 list, and {JUQUEEN},
a Blue Gene/Q system ranked as number 5. We reach resolutions with more
than one trillion cells and perform up to 1.93 trillion cell updates per
second using 1.8 million threads. The design and implementation of
wa{LB}erla is driven by a careful analysis of the performance on current
petascale supercomputers. Our fully distributed data structures and
algorithms allow for efficient, massively parallel simulations on these
machines. Elaborate node level optimizations and vectorization using {SIMD}
instructions result in highly optimized compute kernels for the single- and
two-relaxation-time {LBM}. Excellent weak and strong scaling is achieved
for a complex vascular geometry of the human coronary tree.},
author = {Godenschwager, Christian and Schornbaum, Florian and Bauer, Martin and
Köstler, Harald and Rüde, Ulrich},
journal = {Proceedings of the International Conference for High Performance Computing,
Networking, Storage and Analysis on - {SC} '13},
pages = {1--12},
year = {2013},
month = {1},
publisher = {{IEEE}},
url = {https://www10.cs.fau.de/publications/papers/2013/Godenschwager_SC13.pdf},
ieee = {6877468},
doi = {10.1145/2503210.2503273},
keywords = {data structures, flow simulation, lattice Boltzmann methods, mainframes,
optimisation, parallel machines, parallel programming},
file = {FULLTEXT:pdfs/000/000/000000254.pdf:PDF}
}
Downloads: 0
{"_id":"wvAEBNC6L4ogq4wLi","bibbaseid":"godenschwager-schornbaum-bauer-kstler-rde-aframeworkforhybridparallelflowsimulationswithatrillioncellsincomplexgeometries-2013","authorIDs":[],"author_short":["Godenschwager, C.","Schornbaum, F.","Bauer, M.","Köstler, H.","Rüde, U."],"bibdata":{"bibtype":"article","type":"article","title":"A framework for hybrid parallel flow simulations with a trillion cells in complex geometries","abstract":"waLBerla is a massively parallel software framework for simulating complex flows with the lattice Boltzmann method (LBM). Performance and scalability results are presented for SuperMUC, the world's fastest x86-based supercomputer ranked number 6 on the Top500 list, and JUQUEEN, a Blue Gene/Q system ranked as number 5. We reach resolutions with more than one trillion cells and perform up to 1.93 trillion cell updates per second using 1.8 million threads. The design and implementation of waLBerla is driven by a careful analysis of the performance on current petascale supercomputers. Our fully distributed data structures and algorithms allow for efficient, massively parallel simulations on these machines. Elaborate node level optimizations and vectorization using SIMD instructions result in highly optimized compute kernels for the single- and two-relaxation-time LBM. Excellent weak and strong scaling is achieved for a complex vascular geometry of the human coronary tree.","author":[{"propositions":[],"lastnames":["Godenschwager"],"firstnames":["Christian"],"suffixes":[]},{"propositions":[],"lastnames":["Schornbaum"],"firstnames":["Florian"],"suffixes":[]},{"propositions":[],"lastnames":["Bauer"],"firstnames":["Martin"],"suffixes":[]},{"propositions":[],"lastnames":["Köstler"],"firstnames":["Harald"],"suffixes":[]},{"propositions":[],"lastnames":["Rüde"],"firstnames":["Ulrich"],"suffixes":[]}],"journal":"Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '13","pages":"1–12","year":"2013","month":"1","publisher":"IEEE","url":"https://www10.cs.fau.de/publications/papers/2013/Godenschwager_SC13.pdf","ieee":"6877468","doi":"10.1145/2503210.2503273","keywords":"data structures, flow simulation, lattice Boltzmann methods, mainframes, optimisation, parallel machines, parallel programming","file":"FULLTEXT:pdfs/000/000/000000254.pdf:PDF","bibtex":"@article{Godenschwager-2013-ID254,\n title = {A framework for hybrid parallel flow simulations with a trillion cells in\n complex geometries},\n abstract = {wa{LB}erla is a massively parallel software framework for simulating\n complex flows with the lattice Boltzmann method ({LBM}). Performance and\n scalability results are presented for Super{MUC}, the world's fastest\n x86-based supercomputer ranked number 6 on the Top500 list, and {JUQUEEN},\n a Blue Gene/Q system ranked as number 5. We reach resolutions with more\n than one trillion cells and perform up to 1.93 trillion cell updates per\n second using 1.8 million threads. The design and implementation of\n wa{LB}erla is driven by a careful analysis of the performance on current\n petascale supercomputers. Our fully distributed data structures and\n algorithms allow for efficient, massively parallel simulations on these\n machines. Elaborate node level optimizations and vectorization using {SIMD}\n instructions result in highly optimized compute kernels for the single- and\n two-relaxation-time {LBM}. Excellent weak and strong scaling is achieved\n for a complex vascular geometry of the human coronary tree.},\n author = {Godenschwager, Christian and Schornbaum, Florian and Bauer, Martin and\n Köstler, Harald and Rüde, Ulrich},\n journal = {Proceedings of the International Conference for High Performance Computing,\n Networking, Storage and Analysis on - {SC} '13},\n pages = {1--12},\n year = {2013},\n month = {1},\n publisher = {{IEEE}},\n url = {https://www10.cs.fau.de/publications/papers/2013/Godenschwager_SC13.pdf},\n ieee = {6877468},\n doi = {10.1145/2503210.2503273},\n keywords = {data structures, flow simulation, lattice Boltzmann methods, mainframes,\n optimisation, parallel machines, parallel programming},\n file = {FULLTEXT:pdfs/000/000/000000254.pdf:PDF}\n}\n\n","author_short":["Godenschwager, C.","Schornbaum, F.","Bauer, M.","Köstler, H.","Rüde, U."],"key":"Godenschwager-2013-ID254","id":"Godenschwager-2013-ID254","bibbaseid":"godenschwager-schornbaum-bauer-kstler-rde-aframeworkforhybridparallelflowsimulationswithatrillioncellsincomplexgeometries-2013","role":"author","urls":{"Paper":"https://www10.cs.fau.de/publications/papers/2013/Godenschwager_SC13.pdf"},"keyword":["data structures","flow simulation","lattice Boltzmann methods","mainframes","optimisation","parallel machines","parallel programming"],"downloads":0},"bibtype":"article","biburl":"http://woowoowoo.com/ideas/test.bib","creationDate":"2021-02-10T01:28:56.976Z","downloads":0,"keywords":["data structures","flow simulation","lattice boltzmann methods","mainframes","optimisation","parallel machines","parallel programming"],"search_terms":["framework","hybrid","parallel","flow","simulations","trillion","cells","complex","geometries","godenschwager","schornbaum","bauer","köstler","rüde"],"title":"A framework for hybrid parallel flow simulations with a trillion cells in complex geometries","year":2013,"dataSources":["gDKyuywTCAxcaYPT7"]}