• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Konferenzschrift
  4. Joint Schedule and Layout Autotuning for Sparse Matrices with Compound Entries on GPUs
 
  • Details
  • Full
Options
2019
Conference Paper
Title

Joint Schedule and Layout Autotuning for Sparse Matrices with Compound Entries on GPUs

Abstract
Large sparse matrices with compound entries, i.e., complex and quaternionic matrices as well as matrices with dense blocks, are a core component of many algorithms in geometry processing, physically based animation, and other areas of computer graphics. We generalize several matrix layouts and apply joint schedule and layout autotuning to improve the performance of the sparse matrix-vector product on massively parallel graphics processing units. Compared to schedule tuning without layout tuning, we achieve speedups of up to 5:5x. In comparison to cuSPARSE, we achieve speedups of up to 4:7x.
Author(s)
Mueller-Roemer, Johannes Sebastian  orcid-logo
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Stork, André
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Fellner, Dieter W.
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Mainwork
24th International Symposium on Vision, Modeling, and Visualization, VMV 2019  
Project(s)
CloudiFacturing  
Funder
European Commission EC  
Conference
International Symposium on Vision, Modeling, and Visualization (VMV) 2019  
DOI
10.2312/vmv.20191324
Language
English
Fraunhofer-Institut für Graphische Datenverarbeitung IGD  
Keyword(s)
  • Lead Topic: Digitized Work

  • Research Line: (Interactive) simulation (SIM)

  • general purpose computation on graphics processing unit (GPGPU)

  • GPU computing

  • linear systems

  • code generation

  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024