Re: writing netCDF-4 file with compression?

From: Yolande Serra <serra_at_nyahnyahspammersnyahnyah>
Date: Fri, 29 Aug 2008 12:49:14 -0700

Hi,

I'm trying to use NCL to calculate the streamfunction for reanalysis
winds. I have code that can do that now, but am wondering about
outputting the streamfunction as a short (like the NCEP/NCAR winds
are stored) so that the files are not so large. I take it there is no
way to do this?

Thanks, Yolande

On Aug 17, 2008, at 2:49 PM, David Brown wrote:

> Hi Jeff,
>
> There is a file option that can be set using the setfileoption
> procedure called "CompressionLevel".
> This allows you to specify the compression-level for data written
> to a NetCDF4 classic file.
>
> Prior to opening the file you would make these calls:
> setfileoption("nc","Format","NetCDF4Classic")
> setfileoption("nc","CompressionLevel",5) ; 0 through 9 possible
>
> This causes all data written to be compressed. There is no way
> currently to selectively activate compression
> on a per-variable basis.
> -dave
>
>
> On Aug 13, 2008, at 4:09 PM, Jeff Van Dorn wrote:
>
>> Hello,
>>
>> I saw in a message on Jan, 28 2008 "about HDF5", that Mary said:
>>
>> "We do support classic netCDF-4 files (on systems where netCDF-4 is
>> supported), which write HDF5 underneath the hood."
>>
>> I'm wondering if you're able to modify the compression of the
>> netCDF file on output (e.g., from the C netCDF API:
>> nc_def_var_deflate)? Or does "classic" mean NCL can write
>> netCDF-4 files but with just using standard settings?
>>
>> Thanks,
>> Jeff
>>
>>
>>
>> _______________________________________________
>> ncl-talk mailing list
>> ncl-talk_at_ucar.edu
>> http://mailman.ucar.edu/mailman/listinfo/ncl-talk
>
> _______________________________________________
> ncl-talk mailing list
> ncl-talk_at_ucar.edu
> http://mailman.ucar.edu/mailman/listinfo/ncl-talk

_______________________________________________
ncl-talk mailing list
ncl-talk_at_ucar.edu
http://mailman.ucar.edu/mailman/listinfo/ncl-talk
Received on Fri Aug 29 2008 - 13:49:14 MDT

This archive was generated by hypermail 2.2.0 : Mon Sep 01 2008 - 06:20:11 MDT