Universität zu Köln

Klawonn, Axel ORCID: 0000-0003-4765-7387, Kühn, Martin Joachim ORCID: 0000-0002-0906-6984 and Rheinbach, Oliver (2019). Parallel adaptive FETI-DP using lightweight asynchronous dynamic load balancing. Technical Report.

Preview

PDF CDS_TR-2019-7.pdf Download (13MB) | Preview

Abstract

A parallel FETI-DP domain decomposition method using an adaptive coarse space is presented. The implementation builds on a recently introduced adaptive FETI-DP approach for elliptic problems in three dimensions and uses small, local eigenvalue problems for faces and, additionally, for a small number of edges. The condition number of the preconditioned operator then satisfies a bound which is independent of coefficient heterogeneities in the problem. The computational cost of the local eigenvalue problems is not negligible, and also a significant load imbalance can be introduced. As a remedy, certain eigenvalue problems are discarded by a theory-guided heuristic strategy, based on the diagonal entries of the stiffness matrices. Additionally, a lightweight pairwise dynamic load balancing strategy is implemented for the eigenvalue problems. The load balancing is supervised by an orchestrating rank using asynchronous point-to-point communication. The resulting method shows good weak and strong scalability up to thousands of cores while fast convergence is obtained even for heterogeneous problems.

Item Type:	Monograph (Technical Report)
Creators:	Creators Email ORCID ORCID Put Code Klawonn, Axel axel.klawonn@uni-koeln.de https://orcid.org/0000-0003-4765-7387 UNSPECIFIED Kühn, Martin Joachim martin.kuehn@uni-koeln.de https://orcid.org/0000-0002-0906-6984 UNSPECIFIED Rheinbach, Oliver oliver.rheinbach@math.tu-freiberg.de UNSPECIFIED UNSPECIFIED
URN:	urn:nbn:de:hbz:38-93682
Series Name at the University of Cologne:	Technical report series. Center for Data and Simulation Science
Volume:	2019,7
Date:	21 February 2019
Language:	English
Faculty:	Central Institutions / Interdisciplinary Research Centers
Divisions:	Weitere Institute, Arbeits- und Forschungsgruppen > Center for Data and Simulation Science (CDS)
Subjects:	Data processing Computer science Mathematics Technology (Applied sciences)
Uncontrolled Keywords:	Keywords Language high performance computing English parallelization English domain decomposition methods English partial differential equations English highly heterogeneous English adaptive coarse spaces English
URI:	http://kups.ub.uni-koeln.de/id/eprint/9368