Number of the records: 1  

Particle-in-cell code and Poisson equation solver parallelisation

  1. 1.
    SYSNO ASEP0438779
    Document TypeA - Abstract
    R&D Document TypeThe record was not marked in the RIV
    R&D Document TypeNení vybrán druh dokumentu
    TitleParticle-in-cell code and Poisson equation solver parallelisation
    Author(s) Podolník, Aleš (UFP-V) ORCID
    Komm, Michael (UFP-V) RID, ORCID
    Dejarnac, Renaud (UFP-V) RID, ORCID
    Pánek, Radomír (UFP-V) RID
    Gunn, J. P. (FR)
    Source TitleSymposium on Plasma Physics and Technology 2014. - Prague : Czech Technical University in Prague, Faculty of Electrical Engineering, 2014
    Number of pages1 s.
    Publication formOnline - E
    ActionSPPT 2014 - 26th Symposium on Plasma Physics and Technology/26./
    Event date16.06.2014-19.06.2014
    VEvent locationPrague
    CountryCZ - Czech Republic
    Event typeEUR
    Languageeng - English
    CountryCZ - Czech Republic
    KeywordsParticle-in-cell ; Poisson equation ; parallelisation
    Subject RIVBL - Plasma and Gas Discharge Physics
    Institutional supportUFP-V - RVO:61389021
    AnnotationNumerical simulations based on PIC technique like the SPICE2 model developed at IPP ASCR are often used in tokamak plasma physics to investigate the interaction of edge plasma with plasma-facing components. The SPICE2 model has been parallelised with the exception of the Poisson equation solver, which considerably slows down the simulations. It is now being upgraded to a parallelised version to be efficient enough to perform more demanding tasks like the ITER tokamak baseline scenario edge plasma, which is characterized by very high density (up to 1020 m-3) and low temperature (1–2 eV).This leads to extremely small Debye length (10-7 m) and enormous grids (thousands of cells in each direction).This in combination with a shallow magnetic field inclination (1 – 2 deg.)results in simulations taking several months to compute. Performance and scaling are compared for different solvers using the Helios IFERC supercomputer in order to choose the optimal candidate for specific conditions.
    WorkplaceInstitute of Plasma Physics
    ContactVladimíra Kebza, kebza@ipp.cas.cz, Tel.: 266 052 975
    Year of Publishing2015
Number of the records: 1  

  This site uses cookies to make them easier to browse. Learn more about how we use cookies.