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

[CONDUIT #EJP-757499]: Delay spikes on CONDUIT and what to do

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: [CONDUIT #EJP-757499]: Delay spikes on CONDUIT and what to do
  • Date: Mon, 17 Aug 2015 14:50:06 -0600
Hi Art,

re:
> I'm checking the latency graphs for CONDUIT and have noticed that we're 
> seeing delays of
> 15-30 minutes during certain times of the day, probably during the 0.25 
> degree GFS
> ingest.  Other top-tier sites appear to be seeing delays as well.

We are seeing delays as well, but ours seem to be less than half of what you 
are seeing
at the worst times (~800s vs ~1800s).

re:
> Is the cause of these delays known?

We do not know the cause for the latencies.  I can say that when we were 
ingesting
the full CONDUIT feed (meaning old plus 0.25 deg GFS) directly from a single 
node
in the NCEP cluster (the node had been removed from the cluster for testing 
purposes),
our latencies were _very_ low.

re:
> I'm wondering if I need to be investigating this on my end or whether
> there's some bottleneck close to NCEP that's beyond our control.

Given the difference between you (PSU) and UCAR, I would assume that
there is something between NCEP and PSU.  Given the higher than expected
latencies we are seeing in our feed from NCEP, there is also something
going on at/near NCEP.

I am CCing Carissa Klemmer on this reply so she is aware of the latency
issue.

re:
> Thanks... Art

Please keep us informed if anything changes on your end (either good or bad).

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: EJP-757499
Department: Support CONDUIT
Priority: Normal
Status: Closed