Back
Benedusi_2023_Else_jcp
Journal article

Scalable matrix-free solver for 3D transfer of polarized radiation in stellar atmospheres

  • Benedusi, Pietro ORCID Simula Research Laboratory, Oslo, Norway ; Euler Institute (EUL), Università della Svizzera italiana, Switzerland
  • Riva, Simone ORCID Euler Institute (EUL), Università della Svizzera italiana, Switzerland ; Istituto Ricerche Solari “Aldo e Cele Daccò” (IRSOL), Locarno-Monti, Switzerland
  • Zulian, Patrick ORCID Euler Institute (EUL), Università della Svizzera italiana, Switzerland
  • Štĕpán, Jiri Astronomical Institute of the Czech Academy of Sciences, Ondˇrejov, Czech Republic
  • Belluzzi, Luca ORCID Istituto Ricerche Solari “Aldo e Cele Daccò” (IRSOL), Locarno-Monti, Switzerland ; Leibniz-Institut für Sonnenphysik (KIS), Freiburg i. Br., Germany ; Euler Institute (EUL), Università della Svizzera italiana, Switzerland
  • Krause, Rolf ORCID Euler Institute (EUL), Università della Svizzera italiana, Switzerland ; Faculty of Mathematics and Informatics, FernUni, Brig, Switzerland
Show more…
  • 2023
Published in:
  • Journal of computational physics. - 2023, vol. 479, p. 112013
Radiative transfer
Polarization
Scattering
Krylov methods
Matrix-free
Parallel computing
English We present an efficient and massively parallel solution strategy for the transfer problem of polarized radiation, for a 3D stationary medium out of local thermodynamic equilibrium. Scattering processes are included accounting for partial frequency redistribution effects. Such a setting is one of the most challenging ones in radiative transfer modeling. The problem is formulated for a two-level atomic model, which allows linearization. The discrete ordinate method alongside an exponential integrator are used for discretization. Efficient solution is obtained with a Krylov method equipped with a tailored physics-based preconditioner. A matrix-free approach results in a lightweight implementation, suited for tackling large problems. Near-optimal strong and weak scalability are obtained with two complementary decompositions of the computational domain. The presented approach made it possible to perform simulations for the Cailine at 4227Å with more than 109degrees of freedom in less than half an hour on massively parallel machines, always converging in a few iterations for the proposed tests.
Collections
Language
  • English
Classification
Physics
License
CC BY
Open access status
hybrid
Identifiers
Persistent URL
https://n2t.net/ark:/12658/srd1326711
Statistics
Document views: 30 File downloads:
  • Benedusi_2023_Else_jcp.pdf: 107