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

Re: 20020425: virtual NetCDF data

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: 20020425: virtual NetCDF data
  • Date: Fri, 26 Apr 2002 08:29:26 -0600
>To: address@hidden
>cc: address@hidden
>From: Zhangfan Xing <address@hidden>
>Subject: virtual data
>Organization: UCAR/Unidata
>Keywords: 200204252155.g3PLtja18355

Hi Zhangfan,

> I would like to know if the current netcdf lib
> supports virtual data, i.e., a netcdf entry file
> serves data spanning multiple physical files.
> It's been mentioned in your FAQ as future plan.
> Has that ever got implemented?

No, not directly.  However as part of the DODS (distributed
Oceanographic Data System) project, we have implemented an
"aggregation server" which makes multiple netCDF files look like one
larger netCDF dataset to its clients.  The aggregation server is
currently in beta testing. The configuration files are expected to
change somewhat, and a 1.0 release will be made by the end of May,
2002. See

 http://www.unidata.ucar.edu/projects/THREDDS/tech/DODSAggServer.html

for more information.  As explained on that page, it currently
supports three kinds of aggregation.

There were always some semantic problems with the original idea of
virtual data that were troublesome.  For example, how could a
dimension be shared between two variables in different datasets?  If
the size of the dimension changed in one of the datasets so that they
were no longer consistent, what should an inquiry of the dimension
size return?  The aggregation server approach is letting us explore
what kinds of aggregations are meaningful and useful and what
constraints are required to avoid such problems.

--Russ

_____________________________________________________________________

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