[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: 20050218: restoring headers in large netcdf files

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: Re: 20050218: restoring headers in large netcdf files
  • Date: Wed, 02 Mar 2005 15:12:19 -0700
>To: address@hidden
>cc: Bjarne B?chmann <address@hidden>,
>cc: Peter Gylling J?rgensen <address@hidden>
>From: Karsten Bolding <address@hidden>
>Subject: restoring headers in large netcdf files
>Organization: Bolding & Burchard Hydrodynamics
>Keywords: 200502190949.j1J9ncv2029855 netCDF large file header restore

Karsten,

I wrote:

> I've just tried this to produce a CDF1 file that's bigger than 2 GiB
> for you to test with NCO.  Put this in "large.cdl":
> 
> netcdf large {
> dimensions:
>       x = 1024;
>       y = 1024; 
>       z = 1024;
>         time = UNLIMITED;
> variables:
>       byte var(time, x, y, z); // 1 GiB per record
>       int time(time);
> data:
>   time = 1, 2, 3;
> }
> 
> Then run
> 
>   ncgen -b large.cdl; ls -l large.nc
> 
> For me this took a little over 3 minutes to produce a file over 3 GiB
> in size:
> 
>   -rw-r--r--   1 russ     ustaff   3221225648 Feb 22 16:47 large.nc
> 
> Now this is a CDF1 (classic) netCDF file, so even older versions of
> NCO should be able to handle it, as long as NCO is built with large
> file system support.  If you attempt the NCO operation 
> 
>   ncks -v time large.nc
> 
> I would think it should work, if NCO is built with large file support.

Having just built NCO 2.9.9 with large file support (I used
--enable-largefiles as a configure argument when building, but that
may be the default), I can now verify that ncks built this way does
work fine on the above 3 GiB netCDF classic format (CDF1) file:

  $ ncks -v time large.nc time.nc
  $ ncdump time.nc
  netcdf time {
  dimensions:
          time = UNLIMITED ; // (3 currently)
  variables:
          int time(time) ;

  // global attributes:
                  :history = "Wed Mar  2 14:58:55 2005: ncks -v time large.nc 
time.nc" ;
  data:

   time = 1, 2, 3 ;
  }

If the ncks you have works the same way on the same file generated by
ncgen, then I think it should work fine on the large CDF1 files you
generated with the netcdf=3.6.0 version library.  If ncks fails on
this file, then you need to upgrade your NCO installation.

I also generated a 64-bit offset (CDF2) netCDF file from the above
large.cdl file, using ncgen:

  $ ncgen -v 2 -b large.cdl; ls -l large.nc
  -rw-r--r--   1 russ     ustaff   3221225656 Mar  2 15:08 large.nc

and tested the same ncks, to make sure it would work with both CDF1
and CDF2 files.  It worked identically, so for ncks (and any other
application compiled with netCDF 3.6), it shouldn't matter whether you
have CDF1 or CDF2 files.  But I would recommend just using CDF1 files
for as long as you can, until applications are upgraded to version
3.6.

--Russ

_____________________________________________________________________

Russ Rew                                         UCAR Unidata Program
address@hidden          http://www.unidata.ucar.edu/staff/russ