On the Robustness and Prospects of Adaptive BDDC Methods for Finite Element Discretizations of Elliptic PDEs with High-Contrast Coefficients

Handle URI:
http://hdl.handle.net/10754/621348
Title:
On the Robustness and Prospects of Adaptive BDDC Methods for Finite Element Discretizations of Elliptic PDEs with High-Contrast Coefficients
Authors:
Zampini, Stefano ( 0000-0002-0435-0433 ) ; Keyes, David E. ( 0000-0002-4052-7224 )
Abstract:
Balancing Domain Decomposition by Constraints (BDDC) methods have proven to be powerful preconditioners for large and sparse linear systems arising from the finite element discretization of elliptic PDEs. Condition number bounds can be theoretically established that are independent of the number of subdomains of the decomposition. The core of the methods resides in the design of a larger and partially discontinuous finite element space that allows for fast application of the preconditioner, where Cholesky factorizations of the subdomain finite element problems are additively combined with a coarse, global solver. Multilevel and highly-scalable algorithms can be obtained by replacing the coarse Cholesky solver with a coarse BDDC preconditioner. BDDC methods have the remarkable ability to control the condition number, since the coarse space of the preconditioner can be adaptively enriched at the cost of solving local eigenproblems. The proper identification of these eigenproblems extends the robustness of the methods to any heterogeneity in the distribution of the coefficients of the PDEs, not only when the coefficients jumps align with the subdomain boundaries or when the high contrast regions are confined to lie in the interior of the subdomains. The specific adaptive technique considered in this paper does not depend upon any interaction of discretization and partition; it relies purely on algebraic operations. Coarse space adaptation in BDDC methods has attractive algorithmic properties, since the technique enhances the concurrency and the arithmetic intensity of the preconditioning step of the sparse implicit solver with the aim of controlling the number of iterations of the Krylov method in a black-box fashion, thus reducing the number of global synchronization steps and matrix vector multiplications needed by the iterative solver; data movement and memory bound kernels in the solve phase can be thus limited at the expense of extra local ops during the setup of the preconditioner. This paper presents an exposition of the BDDC algorithm that identifies the current computational bottlenecks that could prevent it from being competitive with other solvers, and proposes solutions in anticipation of exascale architectures. Furthermore, the discussion aims to give interested practitioners sufficient insights to decide whether or not to pursue BDDC in their applications. In addition, the work presents novel numerical results using the distributed memory implementation of BDDC in the PETSc library for vector field problems arising in the context of porous media ows and electromagnetic modeling; the results provide evidence of the robustness of these methods for highly heterogenous problems and non-conforming discretizations. © 2016 ACM.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Extreme Computing Research Center
Citation:
Zampini S, Keyes DE (2016) On the Robustness and Prospects of Adaptive BDDC Methods for Finite Element Discretizations of Elliptic PDEs with High-Contrast Coefficients. Proceedings of the Platform for Advanced Scientific Computing Conference on ZZZ - PASC ’16. Available: http://dx.doi.org/10.1145/2929908.2929919.
Publisher:
Association for Computing Machinery (ACM)
Journal:
Proceedings of the Platform for Advanced Scientific Computing Conference on ZZZ - PASC '16
Conference/Event name:
3rd Conference on Platform for Advanced Scientific Computing, PASC 2016
Issue Date:
2-Jun-2016
DOI:
10.1145/2929908.2929919
Type:
Conference Paper
Appears in Collections:
Conference Papers; Extreme Computing Research Center; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorZampini, Stefanoen
dc.contributor.authorKeyes, David E.en
dc.date.accessioned2016-11-03T06:58:14Z-
dc.date.available2016-11-03T06:58:14Z-
dc.date.issued2016-06-02en
dc.identifier.citationZampini S, Keyes DE (2016) On the Robustness and Prospects of Adaptive BDDC Methods for Finite Element Discretizations of Elliptic PDEs with High-Contrast Coefficients. Proceedings of the Platform for Advanced Scientific Computing Conference on ZZZ - PASC ’16. Available: http://dx.doi.org/10.1145/2929908.2929919.en
dc.identifier.doi10.1145/2929908.2929919en
dc.identifier.urihttp://hdl.handle.net/10754/621348-
dc.description.abstractBalancing Domain Decomposition by Constraints (BDDC) methods have proven to be powerful preconditioners for large and sparse linear systems arising from the finite element discretization of elliptic PDEs. Condition number bounds can be theoretically established that are independent of the number of subdomains of the decomposition. The core of the methods resides in the design of a larger and partially discontinuous finite element space that allows for fast application of the preconditioner, where Cholesky factorizations of the subdomain finite element problems are additively combined with a coarse, global solver. Multilevel and highly-scalable algorithms can be obtained by replacing the coarse Cholesky solver with a coarse BDDC preconditioner. BDDC methods have the remarkable ability to control the condition number, since the coarse space of the preconditioner can be adaptively enriched at the cost of solving local eigenproblems. The proper identification of these eigenproblems extends the robustness of the methods to any heterogeneity in the distribution of the coefficients of the PDEs, not only when the coefficients jumps align with the subdomain boundaries or when the high contrast regions are confined to lie in the interior of the subdomains. The specific adaptive technique considered in this paper does not depend upon any interaction of discretization and partition; it relies purely on algebraic operations. Coarse space adaptation in BDDC methods has attractive algorithmic properties, since the technique enhances the concurrency and the arithmetic intensity of the preconditioning step of the sparse implicit solver with the aim of controlling the number of iterations of the Krylov method in a black-box fashion, thus reducing the number of global synchronization steps and matrix vector multiplications needed by the iterative solver; data movement and memory bound kernels in the solve phase can be thus limited at the expense of extra local ops during the setup of the preconditioner. This paper presents an exposition of the BDDC algorithm that identifies the current computational bottlenecks that could prevent it from being competitive with other solvers, and proposes solutions in anticipation of exascale architectures. Furthermore, the discussion aims to give interested practitioners sufficient insights to decide whether or not to pursue BDDC in their applications. In addition, the work presents novel numerical results using the distributed memory implementation of BDDC in the PETSc library for vector field problems arising in the context of porous media ows and electromagnetic modeling; the results provide evidence of the robustness of these methods for highly heterogenous problems and non-conforming discretizations. © 2016 ACM.en
dc.publisherAssociation for Computing Machinery (ACM)en
dc.subjectAdaptive coarse spacesen
dc.titleOn the Robustness and Prospects of Adaptive BDDC Methods for Finite Element Discretizations of Elliptic PDEs with High-Contrast Coefficientsen
dc.typeConference Paperen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentExtreme Computing Research Centeren
dc.identifier.journalProceedings of the Platform for Advanced Scientific Computing Conference on ZZZ - PASC '16en
dc.conference.date8 June 2016 through 10 June 2016en
dc.conference.name3rd Conference on Platform for Advanced Scientific Computing, PASC 2016en
kaust.authorZampini, Stefanoen
kaust.authorKeyes, David E.en
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.