Implementing the conjugate gradient algorithm on multi-core systems

W.A. Wiggers, Vincent Bakker, Andre B.J. Kokkeler, Gerardus Johannes Maria Smit

    Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

    18 Citations (Scopus)
    109 Downloads (Pure)

    Abstract

    In linear solvers, like the conjugate gradient algorithm, sparse-matrix vector multiplication is an important kernel. Due to the sparseness of the matrices, the solver runs relatively slow. For digital optical tomography (DOT), a large set of linear equations have to be solved which currently takes in the order of hours on desktop computers. Our goal was to speed up the conjugate gradient solver. In this paper we present the results of applying multiple optimization techniques and exploiting multi-core solutions offered by two recently introduced architectures: Intel’s Woodcrest general purpose processor and NVIDIA’s G80 graphical processing unit. Using these techniques for these architectures, a speedup of a factor three has been achieved.
    Original languageUndefined
    Title of host publicationProceedings of the International Symposium on System-on-Chip (SoC 2007)
    EditorsJ. Nurmi, J. Takala, O. Vainio
    Place of PublicationPiscataway, NJ
    PublisherIEEE
    Pages11-14
    Number of pages4
    ISBN (Print)1-4244-1367-2
    DOIs
    Publication statusPublished - 20 Nov 2007
    EventInternational Symposium on System-on-Chip, SoC - Tampere, Finland
    Duration: 19 Nov 200721 Nov 2007

    Publication series

    Name
    PublisherIEEE
    Number07ex1846

    Conference

    ConferenceInternational Symposium on System-on-Chip, SoC
    Period19/11/0721/11/07
    Other19-21 Nov 2007

    Keywords

    • CAES-EEA: Efficient Embedded Architectures
    • METIS-245791
    • EWI-11441
    • IR-64476

    Cite this