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

[IDD #WFA-619667]: Request for CONDUIT feed

Hi Randy and Martha,

Here is another "smoking gun" piece of information that shows that
there is definitely something going on with the data flow from
idd.cise-nsf.gov to noaapxcd.it-protect.com:

I am running simultaneous instances of

'notifyme -vl- -f CONDUIT -o 1 -h idd.cise-nsf.gov'

one from noaapxcd.it-protect.com and another from yakov.unidata.ucar.edu.

The 'notifyme' invocation from yakov to idd.cise-nsf.gov is showing
CONDUIT data rapidly being put into the LDM queue on idd.cise-nsf.gov.
And, the time difference between the product insertion time and the
'notifyme' log time are negligible:

yakov's notifyme to idd.cise.nsf.gov:

 ...

Feb 18 22:38:45 notifyme[4216] INFO:    31502 20090218223842.847 CONDUIT 014  
data/nccf/com/gens/prod/gefs.20090218/18/pgrb2a/gep17.t18z.pgrb2af00 
!grib2/ncep/SPEC62MRF/#000/200902181800F000/RELH/700 Pa PRES! 000014
Feb 18 22:38:45 notifyme[4216] INFO:    20999 20090218223842.834 CONDUIT 005  
data/nccf/com/gens/prod/gefs.20090218/18/pgrb2a/gep09.t18z.pgrb2af00 
!grib2/ncep/SPEC62MRF/#000/200902181800F000/TMPK/250 Pa PRES! 000005
Feb 18 22:38:45 notifyme[4216] INFO:   117335 20090218223842.848 CONDUIT 015  
data/nccf/com/gens/prod/gefs.20090218/18/pgrb2a/gep17.t18z.pgrb2af00 
!grib2/ncep/SPEC62MRF/#000/200902181800F000/UREL;VREL/700 Pa PRES! 000015
Feb 18 22:38:45 notifyme[4216] INFO:   117368 20090218223842.856 CONDUIT 015  
data/nccf/com/gens/prod/gefs.20090218/18/pgrb2a/gep09.t18z.pgrb2af00 
!grib2/ncep/SPEC62MRF/#000/200902181800F000/UREL;VREL/700 Pa PRES! 000015

 ...

The delta(time) for these products are:

20090218223845 - 20090218223842.847 = 2.153 seconds
20090218223845 - 20090218223842.834 = 2.166 seconds
20090218223845 - 20090218223842.848 = 2.152 seconds
20090218223845 - 20090218223842.856 = 2.144 seconds

At the same time, the 'notifyme' run from noaapxcd to idd.cise-nsf.gov
shows large delta(time)s.  In fact, the longer that the 'notifyme' runs,
the bigger the time difference gets.  You can see this effect by doing
the following:

<as 'ldm' on noaapxcd>
notifyme -vl- -f CONDUIT -o 1 -h idd.cise-nsf.gov

This invocation essentially says to report the products just being
inserted into the LDM queue on idd.cise-nsf.gov.  If you leave this
invocation running, you will see the delta(time)s increase with time:
it starts in the few second range and then increases with each product
listed. The same invocation run on my workstation here at the UPC shows
that the delta(time) is essentially fixed at about 2 seconds.

This demonstrates that there is something slowing down the receipt of
data being sent from idd.cise-nsf.gov to noaapxcd.it-protect.com. We
suspect packet shaping or detailed/deep packet inspection somewhere
in NGC.  If this can not be turned off or sped up, your chance of
getting the CONDUIT data you want via the IDD is slim.

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: WFA-619667
Department: Support IDD
Priority: Normal
Status: Closed