Parallel multigrid on hierarchical hybrid grids: a performance study on current high performance computing clusters

SUMMARY This article studies the performance and scalability of a geometric multigrid solver implemented within the hierarchical hybrid grids (HHG) software package on current high performance computing clusters up to nearly 300,000 cores. HHG is based on unstructured tetrahedral finite elements tha...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Concurrency and computation Jg. 26; H. 1; S. 217 - 240
Hauptverfasser: Gmeiner, Björn, Köstler, Harald, Stürmer, Markus, Rüde, Ulrich
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Blackwell Publishing Ltd 01.01.2014
Schlagworte:
Clusters
finite elements
HHG
HPC cluster
Jugene
multigrid
performance model
ISSN:1532-0626, 1532-0634
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:SUMMARY This article studies the performance and scalability of a geometric multigrid solver implemented within the hierarchical hybrid grids (HHG) software package on current high performance computing clusters up to nearly 300,000 cores. HHG is based on unstructured tetrahedral finite elements that are regularly refined to obtain a block‐structured computational grid. One challenge is the parallel mesh generation from an unstructured input grid that roughly approximates a human head within a 3D magnetic resonance imaging data set. This grid is then regularly refined to create the HHG grid hierarchy. As test platforms, a BlueGene/P cluster located at Jülich supercomputing center and an Intel Xeon 5650 cluster located at the local computing center in Erlangen are chosen. To estimate the quality of our implementation and to predict runtime for the multigrid solver, a detailed performance and communication model is developed and used to evaluate the measured single node performance, as well as weak and strong scaling experiments on both clusters. Thus, for a given problem size, one can predict the number of compute nodes that minimize the overall runtime of the multigrid solver. Overall, HHG scales up to the full machines, where the biggest linear system solved on Jugene had more than one trillion unknowns. Copyright © 2012 John Wiley & Sons, Ltd.
Bibliographie:This article is part of the Special issue on High-performance computing and simulation (HPCS 2011) published in Concurrency and Computation: Practice and Experience, Volume 27, Issue 9. The entire Special Issue can be read on Wiley Online Library at: http://onlinelibrary.wiley.com/doi/10.1002/cpe.v27.9/issuetoc
istex:D3A9AA393776575FF53659D94E6765ED80EBF29E
ArticleID:CPE2968
ark:/67375/WNG-T4CWH4TX-0
ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:1532-0626
1532-0634
DOI:10.1002/cpe.2968