SPECFEM / specfem3d

SPECFEM3D_Cartesian simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra (structured or not).
GNU General Public License v3.0
412 stars 229 forks source link

CUDA support for seismograms #1032

Closed EtienneBachmann closed 7 years ago

EtienneBachmann commented 7 years ago

Hi all,

In the next days, I'll add the CUDA support for seismograms compution, based on the way it is implemented on the 2D version. I will try to extend it to fluid-solid media ( in 2D the code currently only stands for pure fluid or pure solid simulations) to fit the possible uses that can be useful for an extent to 3D_globe. Nevertheless, it can be quite long for me to implement computation of pressure seismo in solids, and displ/veloc/accel seismo in fluid media. Is it OK if I put a stop statement in the GPU code stating that computing pressure is only possible for a seismo located in fluids and displ for seismo located in solids ?

Best regards,

Etienne

komatits commented 7 years ago

Hi Etienne, Hi all,

Yes, I think it is fully OK to put some stop statements for now. This is a classical way of adding features in SPECFEM: someone adds a few cases, and stop statements for the other cases, and a few months / years later someone adds the missing cases. Works fine :-)

I have added your name to the list of authors of the code ;-)

Best wishes, Dimitri.

On 06/08/2017 03:13 AM, EtienneBachmann wrote:

Hi all,

In the next days, I'll add the CUDA support for seismograms compution, based on the way it is implemented on the 2D version. I will try to extend it to fluid-solid media ( in 2D the code currently only stands for pure fluid or pure solid simulations) to fit the possible uses that can be useful for an extent to 3D_globe. Nevertheless, it can be quite long for me to implement computation of pressure seismo in solids, and displ/veloc/accel seismo in fluid media. Is it OK if I put a stop statement in the GPU code stating that computing pressure is only possible for a seismo located in fluids and displ for seismo located in solids ?

Best regards,

Etienne

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/geodynamics/specfem3d/issues/1032, or mute the thread https://github.com/notifications/unsubscribe-auth/AFjDKVmxCzRlb6ittFmKKkTpUAjI4Gbiks5sB0rIgaJpZM4Nzdof.

-- Dimitri Komatitsch, CNRS Research Director (DR CNRS) Laboratory of Mechanics and Acoustics, Marseille, France http://komatitsch.free.fr

EtienneBachmann commented 7 years ago

Perfect! The job is done, this issue can be closed.

komatits commented 7 years ago

Thanks a lot! Closed.