Block-asynchronous multigrid smoothers for GPU-accelerated systems

Hartwig Anzt, Stanimire Tomov, Mark Gates, Jack Dongarra, Vincent Heuveline

Research output: Contribution to journalConference articlepeer-review

14 Scopus citations

Abstract

This paper explores the need for asynchronous iteration algorithms as smoothers in multigrid methods. The hardware target for the new algorithms is top-of-the-line, highly parallel hybrid architectures - multicore-based systems enhanced with GPGPUs. These architectures are the most likely candidates for future high-end supercomputers. To pave the road for their efficient use, we must resolve challenges related to the fact that data movement, not floatingpoint operations, is the bottleneck to performance. Our work is in this direction - we designed block-asynchronous multigrid smoothers that perform more flops in order to reduce synchronization, and hence data movement. We show that the extra flops are done for "free," while synchronization is reduced and the convergence properties of multigrid with classical smoothers like Gauss-Seidel can be preserved.

Original languageEnglish
Pages (from-to)7-16
Number of pages10
JournalProcedia Computer Science
Volume9
DOIs
StatePublished - 2012
Event12th Annual International Conference on Computational Science, ICCS 2012 - Omaha, NB, United States
Duration: Jun 4 2012Jun 6 2012

Keywords

  • Block-asynchronous iteration
  • GPU
  • Multigrid smoothers

Fingerprint

Dive into the research topics of 'Block-asynchronous multigrid smoothers for GPU-accelerated systems'. Together they form a unique fingerprint.

Cite this