Bibliographic Metadata

Title
Efficient setup of aggregation AMG for CFD on GPUs
AuthorEmans, Maximilian In der Gemeinsamen Normdatei der DNB nachschlagen ; Liebmann, Manfred
Published in
Applied Parallel and Scientific Computing / Manninen, Pekka; Öster, Per, Heidelberg [u.a.], 2013, page 398-409
Published2013
Edition
Accepted version
LanguageEnglish
SeriesLecture Notes in Computer Science ; 7782
Document typeArticle in a collected edition
Keywords (EN)AMG / Multigrid
ISBN978-3-642-36803-5
URNurn:nbn:at:at-ubg:3-4070 Persistent Identifier (URN)
DOI10.1007/978-3-642-36803-5_29 
Restriction-Information
 The work is publicly available
Files
Efficient setup of aggregation AMG for CFD on GPUs [0.2 mb]
Links
Reference
Classification
Abstract (English)

We explore a GPU implementation of a Krylov-accelerated algebraic multigrid (AMG) algorithm with flexible preconditioning. We demonstrate by means of two benchmarks from an industrial computational fluid dynamics (CFD) application that the acceleration with multiple graphics processing units (GPUs) speeds up the solution phase by a factor of up to 13. In order to achieve good performance for the whole AMG algorithm, we propose for the setup a substitution of the double-pairwise aggregation by a simpler aggregation scheme skipping the calculation of temporary grids and operators. The version with the revised setup reduces the total computing time on multiple GPUs by further 30% compared to the GPU implementation with the double-pairwise aggregation. We observe that the GPU implementation of the entire Krylov-accelerated AMG runs up to four times faster than the fastest central processing unit (CPU) implementation.

Stats
The PDF-Document has been downloaded 26 times.