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

[CONDUIT #IPC-356325]: Any idea what's up with the large lags on CONDUIT the past several days?

Hi Pete,

This is a follow-up to the email that Art Person sent yesterday
about his changing from a two-way to a twenty-way split of his
CONDUIT feed REQUESTs:

I implemented a twenty-way split on the Unidata CONDUIT downstream
(conduit.unidata.ucar.edu), and it did result in significantly lower
latencies overnight, but not nearly as dramatically as we had hoped.
Looking at the latency plot for your machine:

http://rtstats.unidata.ucar.edu/cgi-bin/rtstats/iddstats_nc?CONDUIT+idd.aos.wisc.edu

gives a hint of suggestion that you too have modified how you are
REQUESTing CONDUIT from NCEP.  Is this true?

As an FYI, here is how I implemented our twenty-way split:

request CONDUIT "[02468]0$" conduit.ncep.noaa.gov
request CONDUIT "[13579]0$" conduit.ncep.noaa.gov
request CONDUIT "[02468]1$" conduit.ncep.noaa.gov
request CONDUIT "[13579]1$" conduit.ncep.noaa.gov
request CONDUIT "[02468]2$" conduit.ncep.noaa.gov
request CONDUIT "[13579]2$" conduit.ncep.noaa.gov
request CONDUIT "[02468]3$" conduit.ncep.noaa.gov
request CONDUIT "[13579]3$" conduit.ncep.noaa.gov
request CONDUIT "[02468]4$" conduit.ncep.noaa.gov
request CONDUIT "[13579]4$" conduit.ncep.noaa.gov
request CONDUIT "[02468]5$" conduit.ncep.noaa.gov
request CONDUIT "[13579]5$" conduit.ncep.noaa.gov
request CONDUIT "[02468]6$" conduit.ncep.noaa.gov
request CONDUIT "[13579]6$" conduit.ncep.noaa.gov
request CONDUIT "[02468]7$" conduit.ncep.noaa.gov
request CONDUIT "[13579]7$" conduit.ncep.noaa.gov
request CONDUIT "[02468]8$" conduit.ncep.noaa.gov
request CONDUIT "[13579]8$" conduit.ncep.noaa.gov
request CONDUIT "[13579]9$" conduit.ncep.noaa.gov
request CONDUIT "[02468]9$" conduit.ncep.noaa.gov

I don't claim that this is the best way of doing such a split,
but it _should_ make sure that there is a good distribution of
number of products and corresponding volumes in each of the
REQUESTs.

On a totally separate topic: did you receive the email that I
(thought) I sent you, Victor and Russ regarding testing of
the revamped NIMAGE feed?  Just curious if you are interested
in kicking the tires on the new NIMAGE feed and its use in
(a current version of) GEMPAK.

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: IPC-356325
Department: Support CONDUIT
Priority: Critical
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.