I'd like to correct a misunderstanding.
> ... I am very concerned that switched to CDF2 only gains us a factor
> of 2. ... If NetCDF doesn't transition allow larger records, in the
> near future we will either have to redesign our output and analysis
> tools, which is time consuming, or we won't be able to use NetCDF.
The CDF2 format (also known as the "64-bit offset" format) has no 4
GiB limit on record size. Records can be much larger than that, as
the table in the Users Guide shows:
In versions 3.6 and later, the theoretical limit on a record size is
the same as the maximum file size, 8 EiB (about 9.22e+18 bytes). With
more than one record in a file, that has to be divided by the number
of records, to not exceed the maximum file size.
However, no record variable can require more than 4 GiB bytes of
storage for each record's worth of data, unless it is the last record
variable. But you can have billions of record variables.
_From what you've described, I don't think the CDF2 format limits will
constrain your move to higher resolutions, assuming you use a
different netCDF record variable for each 3D field. I'm sorry if the
documentation seems to imply a 4 GiB record size limit, but it's
actually a 4 GiB per variable record size limit. If you find you
can't store very large records, that's a bug and we'd like to hear
To unsubscribe netcdfgroup, visit: