We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

math.NA

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Mathematics > Numerical Analysis

Title: Auto-Tuned Preconditioners for the Spectral Element Method on GPUs

Abstract: The Poisson pressure solve resulting from the spectral element discretization of the incompressible Navier-Stokes equation requires fast, robust, and scalable preconditioning. In the current work, a parallel scaling study of Chebyshev-accelerated Schwarz and Jacobi preconditioning schemes is presented, with special focus on GPU architectures, such as OLCF's Summit. Convergence properties of the Chebyshev-accelerated schemes are compared with alternative methods, such as low-order preconditioners combined with algebraic multigrid. Performance and scalability results are presented for a variety of preconditioner and solver settings. The authors demonstrate that Chebyshev-accelerated-Schwarz methods provide a robust and effective smoothing strategy when using $p$-multigrid as a preconditioner in a Krylov-subspace projector. At the same time, optimal preconditioning parameters can vary for different geometries, problem sizes, and processor counts. This variance motivates the development of an autotuner to optimize solver parameters on-line, during the course of production simulations.
Comments: 12 pages, 6 figures, submitted to SIAM Conference on Parallel Processing for Scientific Computing (PP22) proceedings
Subjects: Numerical Analysis (math.NA); Performance (cs.PF)
MSC classes: 35-04
ACM classes: D.0; F.2; G.1.3; G.4; I.6
Cite as: arXiv:2110.07663 [math.NA]
  (or arXiv:2110.07663v1 [math.NA] for this version)

Submission history

From: Malachi Phillips [view email]
[v1] Thu, 14 Oct 2021 18:47:54 GMT (6251kb,D)

Link back to: arXiv, form interface, contact.