Jeff Whitaker <jswhit@xxxxxxxxxxx> writes:
> For what it's worth, I'm also seeing very poor performance writing
> large chunks of data along the unlimited dimension in NETCDF4 and
> NETCDF4_CLASSIC (using the python interface, which is just a wrapper
> around the C interface). Works fine if the format is NETCDF3 though.
> So, I don't think it's an issue with the fortran interface.
Yes, I understand that.
I suspect that this will either be an HDF5 bug, or, more likely, that
we are just doing something slightly incorrectly. (For example not
setting our chunking parameters correctly.)
I will reproduce this problem in HDF5 only (i.e. without netCDF) and
see what the HDF5 programmers have to say about it.
Ed Hartnett -- ed@xxxxxxxxxxxxxxxx
To unsubscribe netcdfgroup, visit: