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

Re: Compressed 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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: Compressed netCDF files
  • Date: Mon, 06 Apr 1998 10:54:38 -0600
> From: address@hidden
> To: Russ Rew <address@hidden>
> Subject: Compressed netCDF files
> Organization:  Northeast Regional Climate Center, Cornell University

Hi Bill,

> Russ -- I have sat on this far too long and want to get it out to people.
> 
> Here is a link to a page making available the compressed netcdf libraries.
> I have been using this code with good success since last summer and just
> recently had the other Climate Centers compile it up on a range of
> architectures with no problems.  Looks like it works on dUNIX, MacOS, linux,
> IRIX, Solaris, AIX, openstep, Rhapsody.
> 
> http://snow.cit.cornell.edu/noon/z_netcdf.html
> 
> The only problem that I need to look into is restructuring the code to
> copy the netcdf file when adding variables or attributes.  Currently it 
> does this very inefficiently and I work around it by de-compressing the file,
> altering the structure, and re-compressing it.
> 
> If you have time, take a look at the changes and compile it up.  If you
> want a compiled library, I can see if I can get one for your OS.
> 
> Looking for comments.  I would like to send this off to the netcdf list
> next week to get others to try it out who use netcdfs differently than
> I do.

Thanks for sending this, it looks quite useful.  I'm CC:ing Glenn Davis
and Steve Emmerson also, in case they have comments.  I notice that you
used netCDF 3.3.1 as a code base.  Would it be practical to integrate
your changes into the latest 3.4 release instead, so users could have
the latest bug fixes and performance improvements?  I know this might be
a *lot* of work, ...

--Russ