Bibliographic Metadata

Title
Steps Towards GPU Accelerated Aggregation AMG
AuthorEmans, Maximilian ; Liebmann, Manfred ; Basara, Branislav
Published in
International Symposium on Parallel and Distributed Computing, New York, 2012, page 79-86
Published2012
LanguageEnglish
Document typeArticle in a collected edition
Keywords (EN)GPU / AMG / algebraic / multigrid / finite / volumes
ISBN978-0-7695-4805-0
URNurn:nbn:at:at-ubg:3-4145 Persistent Identifier (URN)
DOI10.1109/ISPDC.2012.19 
Restriction-Information
 The work is publicly available
Files
Steps Towards GPU Accelerated Aggregation AMG [0.17 mb]
Links
Reference
Classification
Abstract (English)

We present an implementation of AMG with simple aggregation techniques on multiple GPUs. It supports the parallel matrix representations typically used for finite volume discretisation. We employ the ICRS sparse matrix format and the asynchronous exchange mechanism of MPI on CPUs that has been modified to make it suitable for the GPU coprocessors. We show that the solution phase of the standard v-cycle AMG with simple aggregation is accelerated by a factor of up to 12. The solution phase of the more advanced Krylov-accelerated AMG runs faster by a factor of up to 7 on Nvidia TESLA C2070 compared to calculation on Intel X5650 CPUs.

Notice
Stats
The PDF-Document has been downloaded 24 times.