ABOUT US

RESEARCH

RESEARCHERS

REPORTS

SOFTWARE

FACILITIES

EMAIL SERVICES

WIKIS

AuthorMichael Rasquin, Cameron Smith, Kedar Chitale, Seegyoung Seol, Benjamin A. Matthews, Jeffrey L. Martin, Onkar Sahni, Raymond M. Loy, Mark S. Shephard, Kenneth E. Jansen
TitleScalable fully implicit finite element flow solver with application to high-fidelity flow control simulations on a realistic wing design
Year2014
JournalComputing in Science and Engineering
AbstractMassively parallel computation provides enormous capacity to perform simulations on a time scale that can change the paradigm of how simulations are used by scientists, engineers and other practitioners to address discovery and design. We consider in this work an active flow control application on a realistic wing design that could be leveraged only by a scal- able fully implicit unstructured nite element flow solver and the access to high performance computing resources. After a brief introduction, we fi rst describe the main objectives and promises of our active flow control application. We then summarize the main features in the implementation of our massively parallel flow solver, which can address turbulent flow phenomena on any arbitrary complex geometries. Finally, we demonstrate its excellent strong scalability at extreme scale. For that purpose, scaling studies are performed with unstructured meshes of 11 and 92 billion elements on the Argonne Leadership Computing Facility's Blue Gene/Q Mira machine with up to 786,432 cores and 3,145,728 MPI processes.
PDF File Download