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

Re: C++ interface and efficient multi-dimensional arrays (compression?)

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: C++ interface and efficient multi-dimensional arrays (compression?)
  • Date: Wed, 06 Jan 1999 09:41:25 -0700
>To: address@hidden, "Glenn P. Davis" <address@hidden>
>Subject: C++ interface and efficient multi-dimensional arrays (compression?)
>Date: Thu, 24 Dec 1998 15:08:59 -0700
>From: Joe Van Andel <address@hidden>

Joe,

> Given that you have to change the version number in the netCDF
> datafiles, I'd strongly lobby for supporting data compression and/or
> multiple unlimited dimensions in netCDF 4, along with packing.  It
> looks like HDF5 supports multiple unlimited dimensions - so will
> netCDF 4 also support this feature?

We hope so, but I can't commit to support for multiple unlimited
dimensions until we actually try to implement this and see what the
real tradeoffs and problems are.

> Given that I don't want to consume lots of extra space for unused data 
> points, but that I want to use something simple like netCDF, do you 
> think that using the zlib patch is a reasonable interim solution, until 
> netCDF4 is available in a year or two?

I think the zlib patch is reasonable for files that are written once
and treated as read-only after that.  I'd hesitate to recommend it for
files that have to be modified, since changing the value of a single
datum can be expensive if it requires recompression.

Also, given our current schedule, it seems unlikely that netCDF 4 will
be available before the end of 2001, and that schedule is also subject
to change with changing priorities in the Unidata program.

--Russ