You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
@jonathanschilling recently updated SPEC to be able to write all the outputs into hdf5 file using Parallel HDF5, which could also be used for STELLOPT. Besides the stellopt.xxx file, we should be able to write the outputs into a hdf5 or netcdf file. This will be extremely useful for post-processing.
PS: This is not an assignment. Just create an issue to raise people's interest. Volunteers are encouraged.
The text was updated successfully, but these errors were encountered:
Just a note, the BEAMS3D and FIELDLINES code already does this. There are some big caveates to doing this correctly. Also if you don't have a parallel filesystem then alternative coding needs to be included. I would suggest the following enhancements to get this to work correctly:
VMEC needs an option where it writes it's quantities to a group in an HDF5 file. Note this function needs to allow the calling routine to specify the Group not just the file that the quantities are written to.
The STELLOPT code needs to be modified so that stellopt.ext becomes a stellopt_ext.h5 file. I this is alot of work. I would suggest that each iteration be considered a group unto itself.
These are the big two issues once their done, most of the work is done for us.
@jonathanschilling recently updated SPEC to be able to write all the outputs into
hdf5
file usingParallel HDF5
, which could also be used forSTELLOPT
. Besides thestellopt.xxx
file, we should be able to write the outputs into a hdf5 or netcdf file. This will be extremely useful for post-processing.PS: This is not an assignment. Just create an issue to raise people's interest. Volunteers are encouraged.
The text was updated successfully, but these errors were encountered: