**26 commits**| Last commit

**≈ 3 months ago**

Ctrl K

GMGPolar is a geometric multigrid solver using implicit extrapolation to raise the convergence order. It is based on meshes in tensor- or product-format. GMGPolar's focus applications are geometries that can be described by polar or curvilinear coordinates for which suited smoothing is available.

5

contributors

Get started

GMGPolar is a performant geometric multigrid solver using implicit extrapolation to raise the convergence order. It is based on meshes in tensor- or product-format. GMGPolar's focus applications are geometries that can be described by polar or curvilinear coordinates for which suited smoothing procedures have been developed. The software offers both, matrix-free and matrix versions of the multigrid scheme.

GMGPolar's main application comes from plasma fusion. Tokamak fusion reactors are promising alternatives for future energy production. Gyrokinetic simulations are important tools to understand physical processes inside tokamaks and

to improve the design of future plants. In gyrokinetic codes such as Gysela, these simulations

involve at each time step the solution of a gyrokinetic Poisson equation defined on disk-like

cross sections. With its node-level parallelism, GMGPolar can efficiently solve a related problem on a Tokamak cross section with a target size between millions and a billion of dofs.

Keywords

Energy

Geometric Multigrid

High performance computing

Modeling

Multigrid

Plasma fusion

Simulation

Supercomputing

Programming languages

- Other 1%

License

- Apache-2.0

</>Source code

MK

MK

Martin J Kühn

German Aerospace Center

CK

Carola Kruse

CERFACS

PL

Philippe Leleux

CS

Christina Schwarz

UR

Ulrich Rüde

FAU Erlangen-Nürnberg