Hello Linda,
As we had discussed we prefer to open up port 388 on a per request basis at
this stage. This is not a problem for us. However, Dov and I just had a
brief discussion with our local "LDM specialist", Réjean Dumas at our end
and I have asked Réjean to contact Steve to clarify a few points.
We all agree that a community announcement will give us a much better
feeling for the number of sites interested in the GEM data. We will send
you an updated text on the data to help craft this annoucement.
Cheers,
Richard.
-----Message d'origine-----
De: Linda Miller [mailto:address@hidden]
Date: 6 mars, 2002 15:29
À: address@hidden
Cc: address@hidden; address@hidden;
address@hidden; address@hidden;
address@hidden
Objet: 20020301: GEM/CMC model data (fwd)
Hi Richard,
I've appended an email exchange on the GEM model and the interest at the
University of Illinois.
As you know, we have not yet sent an official announcement to the Unidata
university community about the GEM data availability. If you read through
the
email below, you'll find that there is an issue of the firewall on your end.
Will CMC be able to feed each university as they request the data, i.e.,
point-to-point directly? If you recall, we were going to begin in that
manner
to see just how much interest there would be among the university sites. If
the interest is high, it might be considered as an additional data feed in
Unidata's IDD structure, where top level sites could feed others. Tt would
be
good to determine the community interest in GEM, before that would happen,
if
possible.
Could you let us know about the firewall issue as soon as possible so we can
get the community announcement created and sent? Unidata is not really in
the
business of providing real-time data to other sites, but we are doing that
now
for the COMET need and now the Univ of Illinois for the GEM data.
Thanks very much!
Linda
---------- Forwarded Message ----------
Date: Friday, March 01, 2002 12:31 PM -0700
From: Steve Chiswell <address@hidden>
To: address@hidden
Subject: 20020301: GEM/CMC model data
David,
You can request the CMC (aka GEM) feedtype from thelma.ucar.edu
from flood, squall, or data (all are in the allow list of thelma).
Just use the ".*" pattern for feedtype CMC.
We are in the process of moving relays from motherlode to thelma, but
have been semi-sidetracked for the past 2 weeks with scsi/disk problems
on both machines. Both appear somewhat stable, but that could change.
My hope is that at some point in the future, sites will be able to feed
directly
from CMC. However, in the process of setting up this feed, we discovered
that
CMC had to open their firewall for port 388 on a per-IP basis. I suppose
Linda will want to talk with them regarding how they will want to deal with
that issue- either open port 388 globally, or on a per request basis.
At any rate, the data is not very large at this time, so adding that to
thelma for relay isn't a big deal.
The following pqact.conf actions may be useful in filing and decoding the
data
(lines will probably be wrapped in the mailer....so beware of tabs and the
long
pattern in the FILE entry:
#
# File the CMC GEM model by forecast hour
CMC <tab>
CMC_GEM_reg_.*_.*_([12][0-9][0-9][0-9])([0-1][0-9])([0-3][0-9])([0-2][0-9])_
(.*
).grib FILE data/cmc/\1\2\3\4_\5.cmcgem
#
# Decode the data into GEMPAK files using dcgrib2
CMC CMC_GEM_reg.*
PIPE decoders/dcgrib2 -v 1 -d data/gempak/logs/dcgrib.log
-e GEMTBL=/home/gempak/NAWIPS/gempak/tables
The data is approximately 367 products totaling 8570204 bytes.
Let me know if you have questions regarding decoding, pqact.conf patterns
etc.
If you need patterns for NOGAPS/COAMPs let me know. I have the dcgrib2
templates
for those in $GEMTBL/grid/gribkey.tbl too, so you can just PIPE
the NOGAPS feed to the decoder. The NOGAPS consists ofa global grid 240,
while there are currently 4 COAMPS regional grids.
Steve Chiswell
Unidata User Support
---------- End Forwarded Message ----------
Linda Miller - address@hidden
External Liaison, Unidata
University Corporation for Atmospheric Research
P.O. Box 3000
Boulder, CO 80307-3000
303 497-8646 fax: 303-497-8690
URL: http://www.unidata.ucar.edu/staff/lmiller/un.act.html
