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

[IDD #BKI-822215]: LDM feed sites

Hi Adam,

re:
> Here is my ldmd:
> 
> #request ANY     ".*"    sirocco.srcc.lsu.edu    PRIMARY
> #request ANY     ".*"    stokes.metr.ou.edu      SECONDARY

OK, first thing:

- you should be requesting data from pavan.srcc.lsu.edu, not sirocco.
  pavan is the IDD relay machine at LSU; sirocco is an internal machine
  that only services internal machines

> request
> EXP|NXRDSRC|NPORT|NIMAGE|GEM|FNMOC|GPS|FAA604|DIFAX|NMC|FSL5|FSL4|FSL3|FSL2|SPARE|UNIDATA
> ".*"   idd.unidata.ucar.edu

Ah Ha!  You are requesting too much data on a single ldmd.conf request.
This is very likely to be one of the things causing high latencies
in your CONDUIT (aka NMC2) and NGRID datafeeds.

> request UNIDATA         ".*"    idd.unidata.ucar.edu
> #
> # Nexrad Level 2 Feed
> #
> request NEXRAD2 ".*"    idd.unidata.ucar.edu
> #request NEXRAD3 ".*"   idd.unidata.ucar.edu
> #
> # NLDN Feeds
> #
> request NLDN     ".*"    striker2.atmos.albany.edu
> #
> # NOGAPS model data
> #
> #request FNMOC  ".*"    usgodae3.fnmoc.navy.mil
> #request FNMOC  ".*"    idd.unidata.ucar.edu
> 
> Our bandwidth here is 1Gig to I2 and NLR.  There is no packet shaping going
> on as I am also one of the admins of our campus's network.

OK.  I think your problem is that you are requesting too much data
on a single feed.

> Here is the strange thing...At first I only had one line (ANY) for
> idd.unidata.  EVERY product had the same strange latency problem.  I then
> pulled Nexrad2 by itself...latency disappeared from it...rest stayed the
> same.  Then I pulled UNIDATA out by itself...those feed's latency
> disappeared.  I was going to try and pull just the Conduit out next and see
> what happens but I wanted to contact you first about it.

Your testing supports the contention that you should split your
data requests.

Here is a first cut at a recommendation for a change in your data requests:

# CONDUIT - high resolution NCEP model data
request CONDUIT ".*" idd.unidata.ucar.edu

# NGRID - high resolution NOAAPort model data
request NGRID   ".*" idd.unidata.ucar.edu

# UNIDATA == HDS|IDS|DDPLUS|UNIWISC:
# HDS        - lower resolution NOAAPort model data
# IDS|DDPLUS - global observational data from NOAAPort
# UNIWISC    - GOES-East/West image sectors
request UNIDATA ".*" idd.unidata.ucar.edu

# NEXRAD2 - Nexrad Level II (full volume scan) radar data
request NEXRAD2 ".*" idd.unidata.ucar.edu

# FNEXRAD - Nexrad Level III composites
# FNMOC   - model data from Fleet Numerical
# FSL2    - wind profiler data from NOAA/GFS
# GEM     - model data from the Canadian Meteorological Center
# NIMAGE  - GOES-East/West image sectors and products from NOAAPort

request FNEXRAD|FNMOC|FSL2|GEM|NIMAGE ".*" idd.unidata.ucar.edu

# LIGHTNING - NLDN lightning data from SUNY Albany
request LIGHTNING ".*" striker2.atmos.albany.edu

Comments:

- separating the CONDUIT request into its own feed may or
  may not solve your latency problem; only testing will tell.
  If it does not drop the latencies to near zero, then you
  should split the request five ways:

request CONDUIT "([09]$)" idd.unidata.ucar.edu
request CONDUIT "([18]$)" idd.unidata.ucar.edu
request CONDUIT "([27]$)" idd.unidata.ucar.edu
request CONDUIT "([36]$)" idd.unidata.ucar.edu
request CONDUIT "([45]$)" idd.unidata.ucar.edu

- one feed that you are not requesting is NEXRAD3 (specific request
  is commented out).  These are Nexrad Level III products from NOAAPort.

  You would get these if your 'ANY ".*"' request to pavan.srcc.lsu.edu
  were uncommented and if pavan has allowed you to request the products,
  or if you uncommented your NEXRAD3 request to idd.unidata.ucar.edu.

- I would venture a guess that you do not actually want the following
  feeds:

  DIFAX  - difax like products from U Wisconsin (do you really want these?)
  FAA604 - inactive
  GPS    - SuomiNet products
  SPARE  - spare used for test products sporadically

  The GPS products are not available from idd.unidata.ucar.edu.

- the following feeds must be requested from NOAA/GFS:

  FSL3, FSL4, FSL5

  The are not available in the IDD.

- you may or may not want products in the EXP feed type

Try redoing your data request to LSU using pavan instead of sirocco.
Try reorganizing your feed requests to idd.unidata.ucar.edu as
outlined above.

Please let us know your results.

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: BKI-822215
Department: Support IDD
Priority: Normal
Status: Closed