[CONDUIT #RFO-442890]: GFS upgrade

Hi Carissa,

Sorry for slow reply...

re: 0.25 GFS data
> This is the first I am hearing about sending the GFS parallel through
> CONDUIT, so it must have been Becky.

???  I am confused.  When you say GFS parallel, are you referring to
the 0.25 GFS or are you referring to something else?

What we would like to be able to do is setup a test feed of the 0.25
deg GFS (the one that is scheduled to go operational in January) so
we can evaluate the impact on bandwidth, decoding and use.

re:
> But no problem, lets see how we would
> make this work. So in College Park I feed 3 systems that are load balanced
> into 1 IP for you to grab. In Boulder I feed 3 systems, and if I am correct
> you have 3 IPs that you grab from.

Correct.

re:
> If I started to feed the para to 1 of
> the Boulder systems could you folks take that IP out of your pool and pull
> the feed to evaluate?

Yes.

re:
> Then we don't have to make any changes on our side
> except to start inserting the data.

Right.  We are looking to do the testing in the least intrusive
way possible.

Cheers,

Tom

> address@hidden> wrote:
> >
> > Hi Carissa,
> >
> > In advance of canvassing the community wrt replacing 0.5 deg GFS with
> > 0.25 deg GFS, we need to run some tests here in the UPC to make sure
> > that our packages (e.g., TDS, GEMPAK, IDV, etc.) can handle the data.
> >
> > I believe that we (either you and I or Becky and I) talked about this
> > previously, and I seem to recall that the offer was that one of the
> > top level CONDUIT relay cluster nodes would be removed from the cluster
> > and temporarily reconfigured to relay the 0.25 deg GFS fields.
> >
> > Questions:
> >
> > - am I remembering correctly that this is the approach that you (NCEP)
> >   promoted previously?
> >
> >   If yes, when could a test feed of the 0.25 deg GFS be setup so we
> >   could start testing?
> >
> >   FYI: the UPC machine that would/will be REQUESTing the 0.25 deg GFS
> >   data is:
> >
> >   128.117.149.100 <-> lead.unidata.ucar.edu
> >
> > - if/when such a test feed is available, would it be possible for the
> >   other Unidata community top level relays for CONDUIT to REQUEST the
> >   data so that they can evaluate the impact on their machines and
> >   network?
> >
> >   The sites I have in mind are the same as currently feeding CONDUIT
> >   from you now (e.g., University of Wisconsin/AOS, Penn State,
> >   University of Illinois, and Texas A&M).
> >
> > Cheers,
> >
> > Tom
> > --
> >
> > ****************************************************************************
> > Unidata User Support                                    UCAR Unidata
> > Program
> > (303) 497-8642                                                 P.O. Box
> > 3000
> > address@hidden                                   Boulder, CO
> > 80307
> >
> > ----------------------------------------------------------------------------
> > Unidata HomePage                       http://www.unidata.ucar.edu
> >
> > ****************************************************************************
> >
> >
> > Ticket Details
> > ===================
> > Ticket ID: RFO-442890
> > Department: Support CONDUIT
> > Priority: Normal
> > Status: Closed
> >
> >
> 
> 

Cheers,

Tom
--
****************************************************************************
Unidata User Support                                    UCAR Unidata Program
(303) 497-8642                                                 P.O. Box 3000
address@hidden                                   Boulder, CO 80307
----------------------------------------------------------------------------
Unidata HomePage                       http://www.unidata.ucar.edu
****************************************************************************


Ticket Details
===================
Ticket ID: RFO-442890
Department: Support CONDUIT
Priority: Normal
Status: Closed