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

[IDD #BDI-646847]: gfs grids on ldm



Hi Rick,

re:
> I'm again looking at global gfs grids coming over CONDUIT.  The numbers
> are recently low. The statistics below are generated using McIDAS
> GRDLIST command. One odd thing is the machine "on-noaaport" is not
> requesting CONDUIT from any "wisc" address. Just
> 
> request CONDUIT "[05]$"  idd.unidata.ucar.edu
> request CONDUIT "[16]$"  idd.unidata.ucar.edu
> request CONDUIT "[27]$"  idd.unidata.ucar.edu
> request CONDUIT "[38]$"  idd.unidata.ucar.edu
> request CONDUIT "[49]$"  idd.unidata.ucar.edu
> request CONDUIT "[32]$"  idd.unidata.ucar.edu
> request CONDUIT "[242]$"  idd.unidata.ucar.edu
> request CONDUIT "[130]$"  idd.unidata.ucar.edu
> 
> SSEC machine model   model date/time total number of grids
> NPS1            GFS     21220   0       78475
> NPSERVE1        GFS     21220   0       78475
> NPSERVE2        GFS     21220   0       78475
> ON-NOAAPORT     GFS     21220   0       78475
> 
> NPS1            GFS     21220   6       96406
> NPSERVE1        GFS     21220   6       90663
> NPSERVE2        GFS     21220   6       89195
> ON-NOAAPORT     GFS     21220   6       96406
> 
> NPS1            GFS     21220   12      96406
> NPSERVE1        GFS     21220   12      94637
> NPSERVE2        GFS     21220   12      88588
> ON-NOAAPORT     GFS     21220   12      95822
> 
> NPS1            GFS     21220   18      85345
> NPSERVE1        GFS     21220   18      79418
> NPSERVE2        GFS     21220   18      80975
> ON-NOAAPORT     GFS     21220   18      85345

A quick check of the volume of CONDUIT data that we are receiving
from an LDM feed at NCEP shows that the 0Z runs for both today and
yesterday were measurably lower than for other forecast hours:

https://rtstats.unidata.ucar.edu/cgi-bin/rtstats/iddstats_vol_nc?CONDUIT+conduit.unidata.ucar.edu

As a double check, I looked at the CONDUIT volume being received
on the aggregator or the IDD relay cluster in UWisc/AOS, idd-agg.aos.wisc.edu,
and its shows less data for the 0Z run today and yesterday also:

https://rtstats.unidata.ucar.edu/cgi-bin/rtstats/iddstats_vol_nc?CONDUIT+idd-agg.aos.wisc.edu

Since both UWisc/AOS and Unidata feed directly from NCEP, this says
that NCEP did not send the same number of grids for the 0Z runs.
We have noticed that NCEP has been having problems with delivering
the full suite of CONDUIT products starting sometime in the spring
(or even earlier) and continuing until the present.  I am a bit
surprised that Pete Pokrandt has not reported the lesser number of
CONDUIT products seen today and yesterday (at a minimum) to 
NCEP...

re:
> While I got your ear, is it better to filter out non-GFS (eventually may
> keep more) grids in ldmd.conf or pqact.conf?

It is better to not REQUEST products you are not going to use since
it will save bandwidth.  Getting all of the data but not acting on
it via pattern-action file action(s) will work, but you've already
used bandwidth that was not needed to be used, and the product residency
time in your LDM queue will have been smaller because it has to store
all of the data that is being REQUESTed.

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: BDI-646847
Department: Support IDD
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.