Massively Parallel Large Scale Stokes Flow Simulation
Gmeiner B, Huber M, John L, Rüde U, Waluga C, Wohlmuth BI (2016)
Publication Language: English
Publication Type: Conference contribution, Original article
Publication year: 2016
Pages Range: 333-341
Conference Proceedings Title: NIC Symposium 2016 Proceedings
Event location: John von Neumann Institute for Computing (NIC), Jülich
ISBN: 978-3-95806-109-5
URI: http://juser.fz-juelich.de/record/280633/files/NIC_Series_48.pdf
Open Access Link: http://juser.fz-juelich.de/record/280633/files/NIC_Series_48.pdf
Abstract
In many applications, physical models consisting of a Stokes-type equation that is coupled to a convection-dominated transport equation play an important role, e.g., in mantle-convection or ice-sheet dynamics. In the iterative treatment of such problems the computational cost is usually dominated by the solution procedure for the Stokes part. Hence, we focus on massively scalable and fast multigrid solvers for the arising saddle point problem. To gain deeper insight into the performance characteristics, we evaluate the multigrid efficiency systematically and address the methodology of algorithmic resilience. Three methods based on the HHG software framework will be presented and are shown to solve FE systems with half a billion unknowns even on standard workstations. On petascale systems they furthermore exhibit excellent scalability. This together with the optimised performance on each node leads to superior supercomputing efficiency. Indefinite systems with up to ten trillion (1013) unknowns can be solved in less than 13 minutes compute time.
Authors with CRIS profile
Björn Gmeiner
Lehrstuhl für Informatik 10 (Systemsimulation) (LSS)
Markus Huber
Lehrstuhl für Informatik 10 (Systemsimulation) (LSS)
Ulrich Rüde
Lehrstuhl für Informatik 10 (Systemsimulation) (LSS)
Involved external institutions
Germany (DE)How to cite
APA:
Gmeiner, B., Huber, M., John, L., Rüde, U., Waluga, C., & Wohlmuth, B.I. (2016). Massively Parallel Large Scale Stokes Flow Simulation. In Binder K, Müller M, Kremer M, Schnurpfeil A (Eds.), NIC Symposium 2016 Proceedings (pp. 333-341). John von Neumann Institute for Computing (NIC), Jülich.
MLA:
Gmeiner, Björn, et al. "Massively Parallel Large Scale Stokes Flow Simulation." Proceedings of the NIC Symposium 2016, John von Neumann Institute for Computing (NIC), Jülich Ed. Binder K, Müller M, Kremer M, Schnurpfeil A, 2016. 333-341.
BibTeX: Download