Karsten,
> > Would it be convenient for you to compile the same model against the
> > previous
> > netCDF version 4.1.3 for comparison of times with the 4.2 release candidate?
>
> I've already done that with similar results.
> NetCDF version: 4.2-rc1 of Jan 18 2012 07:56:30 $
> real 0m42.843s
> user 0m3.680s
> sys 0m0.130s
>
> NetCDF version: "3.6.3" of Jan 18 2012 08:53:30 $
> real 0m1.788s
> user 0m1.770s
> sys 0m0.000s
>
> NetCDF version: 4.1.3 of Dec 9 2011 10:42:04 $
> real 0m42.423s
> user 0m3.820s
> sys 0m0.080s
>
> version information printed from executable.
Thanks, that's good to know. It tells us two things:
- we haven't significantly slowed things down with changes from 4.1.3 to 4.2,
and
- the performance problem you're seeing is specific to your model, as others
haven't reported such a problem with version 4.1.x
I'll have to stufy your perf output, but in the meantime, could you provide the
complete
output of
ncdump -s -h
for your file *using the 4.2 version of ncdump* rather than the older installed
version
you apparently used, which showed no information about chunk sizes.
Thanks!
--Russ
Russ Rew UCAR Unidata Program
address@hidden http://www.unidata.ucar.edu
Ticket Details
===================
Ticket ID: AHZ-822837
Department: Support netCDF
Priority: Normal
Status: Closed
NOTE: All email exchanges with Unidata User Support are recorded in the Unidata inquiry tracking system and then made publicly available through the web. If you do not want to have your interactions made available in this way, you must let us know in each email you send to us.