High performance shallow water kernels for parallel overland flow simulations based on FullSWOF2D

Handle URI:
http://hdl.handle.net/10754/624957
Title:
High performance shallow water kernels for parallel overland flow simulations based on FullSWOF2D
Authors:
Wittmann, Roland ( 0000-0002-2290-1876 ) ; Bungartz, Hans-Joachim; Neumann, Philipp
Abstract:
We describe code optimization and parallelization procedures applied to the sequential overland flow solver FullSWOF2D. Major difficulties when simulating overland flows comprise dealing with high resolution datasets of large scale areas which either cannot be computed on a single node either due to limited amount of memory or due to too many (time step) iterations resulting from the CFL condition. We address these issues in terms of two major contributions. First, we demonstrate a generic step-by-step transformation of the second order finite volume scheme in FullSWOF2D towards MPI parallelization. Second, the computational kernels are optimized by the use of templates and a portable vectorization approach. We discuss the load imbalance of the flux computation due to dry and wet cells and propose a solution using an efficient cell counting approach. Finally, scalability results are shown for different test scenarios along with a flood simulation benchmark using the Shaheen II supercomputer.
Citation:
Wittmann R, Bungartz H-J, Neumann P (2017) High performance shallow water kernels for parallel overland flow simulations based on FullSWOF2D. Computers & Mathematics with Applications 74: 110–125. Available: http://dx.doi.org/10.1016/j.camwa.2017.01.005.
Publisher:
Elsevier BV
Journal:
Computers & Mathematics with Applications
Issue Date:
25-Jan-2017
DOI:
10.1016/j.camwa.2017.01.005
Type:
Article
ISSN:
0898-1221
Sponsors:
We thank for financial support by the Kompetenznetzwerk für Wissenschaftliches Höchstleistungsrechnen in Bayern (KONWIHR) for the Multicore-Software-Initiative with the project “Optimization of a multi-functional shallow water solver for complex overland flows” (KONWIHR-IV) and KAUST Supercomputing Laboratory for providing the access to supercomputer Shaheen 2 (project k1050). We thank Ralf-Peter Mundani and Florian Mintgen from the Department of Civil, Geo and Environmental Engineering of the Technical University of Munich for access to the Glasgow scenario datasets as well as for the fruitful discussions on this scenario.
Appears in Collections:
Publications Acknowledging KAUST Support

Full metadata record

DC FieldValue Language
dc.contributor.authorWittmann, Rolanden
dc.contributor.authorBungartz, Hans-Joachimen
dc.contributor.authorNeumann, Philippen
dc.date.accessioned2017-06-12T13:52:07Z-
dc.date.available2017-06-12T13:52:07Z-
dc.date.issued2017-01-25en
dc.identifier.citationWittmann R, Bungartz H-J, Neumann P (2017) High performance shallow water kernels for parallel overland flow simulations based on FullSWOF2D. Computers & Mathematics with Applications 74: 110–125. Available: http://dx.doi.org/10.1016/j.camwa.2017.01.005.en
dc.identifier.issn0898-1221en
dc.identifier.doi10.1016/j.camwa.2017.01.005en
dc.identifier.urihttp://hdl.handle.net/10754/624957-
dc.description.abstractWe describe code optimization and parallelization procedures applied to the sequential overland flow solver FullSWOF2D. Major difficulties when simulating overland flows comprise dealing with high resolution datasets of large scale areas which either cannot be computed on a single node either due to limited amount of memory or due to too many (time step) iterations resulting from the CFL condition. We address these issues in terms of two major contributions. First, we demonstrate a generic step-by-step transformation of the second order finite volume scheme in FullSWOF2D towards MPI parallelization. Second, the computational kernels are optimized by the use of templates and a portable vectorization approach. We discuss the load imbalance of the flux computation due to dry and wet cells and propose a solution using an efficient cell counting approach. Finally, scalability results are shown for different test scenarios along with a flood simulation benchmark using the Shaheen II supercomputer.en
dc.description.sponsorshipWe thank for financial support by the Kompetenznetzwerk für Wissenschaftliches Höchstleistungsrechnen in Bayern (KONWIHR) for the Multicore-Software-Initiative with the project “Optimization of a multi-functional shallow water solver for complex overland flows” (KONWIHR-IV) and KAUST Supercomputing Laboratory for providing the access to supercomputer Shaheen 2 (project k1050). We thank Ralf-Peter Mundani and Florian Mintgen from the Department of Civil, Geo and Environmental Engineering of the Technical University of Munich for access to the Glasgow scenario datasets as well as for the fruitful discussions on this scenario.en
dc.publisherElsevier BVen
dc.subjectHigh performance kernelsen
dc.subjectOverland flowsen
dc.subjectShallow water equationsen
dc.subjectVectorizationen
dc.titleHigh performance shallow water kernels for parallel overland flow simulations based on FullSWOF2Den
dc.typeArticleen
dc.identifier.journalComputers & Mathematics with Applicationsen
dc.contributor.institutionTechnical University of Munich, Boltzmannstraße 3, 85748 Garching, Germanyen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.