PmWiki
Software

DIOGE Ne S

DIOGENeS - DIscOntinuous GalErkin Nanoscale Solvers

DIOGENeS is a software suite dedicated to the numerical modeling of light interaction with nanometer scale structures with applications to nanophotonics and nanoplasmonics. Although the team is already working on several software components that will ultimately be part of this software suite, the development of DIOGENeS will officially start in January 2015 in the context of a Software Development Action supported by Inria (Direction of Technological Development).

DIOGENeS essentially relies on a two layer architecture. The core of the suite is a library of generic software components (data structures and algorithms) for the implementation of high order DG and HDG schemes formulated on unstructured tetrahedral and hybrid structured/unstructured (cubic/tetrahedral) meshes. This library will be used to develop dedicated simulation software for time-domain and frequency-domain problems relevant to nanophotonics and nanoplasmonics, considering various material models.

DIOGENeS is programmed in Fortran 2003 and the underlying algorithms are adapted to distributed memory (MPI) and shared memory (OpenMP) parallel computing.

With version V1.0 of this suite, the first dedicated simulation software relying on the core library of generic components will be a 3D time-domain Maxwell solver able to deal with local dispersion models. This solver will be based on a nodal discontinuous Galerkin (DG) method formulated on a fully unstructured tetrahedral mesh.

Features of version V1.0 (currently planned for the end of 2016)

  • Time-domain Maxwell equations in mixed form
  • Drude, Drude-Lorentz and generalized dispersion models
  • Linear isotropic and anisotropic media
  • Affine and curvilinear tetrahedral elements
  • Nodal DG schemes based on centered or upwind numerical fluxes
  • Arbitrary high order nodal (Lagrange) interpolation of the field components within a mesh cell
  • Explicit time-stepping schemes: 2nd and 4th order leap-frog, and optimized low storage Runge-Kutta schemes
  • Silver-Muller absorbing boundary condition and PML