A Fast GPU-accelerated Mixed-precision Strategy for Fully NonlinearWater Wave Computations

Stefan Lemvig Glimberg, Allan Peter Engsig-Karup, Morten G. Madsen

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    267 Downloads (Pure)

    Abstract

    We present performance results of a mixed-precision strategy developed to improve a recently developed massively parallel GPU-accelerated tool for fast and scalable simulation of unsteady fully nonlinear free surface water waves over uneven depths (Engsig-Karup et.al. 2011). The underlying wave model is based on a potential flow formulation, which requires efficient solution of a Laplace problem at large-scales. We report recent results on a new mixed-precision strategy for efficient iterative high-order accurate and scalable solution of the Laplace problem using a multigrid-preconditioned defect correction method. The improved strategy improves the performance by exploiting architectural features of modern GPUs for mixed precision computations and is tested in a recently developed generic library for fast prototyping of PDE solvers. The new wave tool is applicable to solve and analyze large-scale wave problems in coastal and offshore engineering.
    Original languageEnglish
    Title of host publicationProceedings of ENUMATH 2011
    Number of pages8
    PublisherUniversity of Leicester
    Publication date2011
    Publication statusPublished - 2011
    EventENUMATH 2012: The European Numerical Mathematics and Advanced Applications Conference - Leicester, United Kingdom
    Duration: 5 Sep 20129 Sep 2012
    http://www2.le.ac.uk/departments/mathematics/research/enumath2011

    Conference

    ConferenceENUMATH 2012
    CountryUnited Kingdom
    CityLeicester
    Period05/09/201209/09/2012
    Internet address

    Fingerprint Dive into the research topics of 'A Fast GPU-accelerated Mixed-precision Strategy for Fully NonlinearWater Wave Computations'. Together they form a unique fingerprint.

    Cite this