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

[IDD #LNL-619813]: Is there a page on how thr CRAFT/NEXRAD2 feed works?

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 Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated.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: [IDD #LNL-619813]: Is there a page on how thr CRAFT/NEXRAD2 feed works?
  • Date: Mon, 20 Feb 2006 07:53:59 -0700
Hi Gilbert,

re: 4 pqact instances to process the Level II images

> Just these: request ".*" CRAFT idd.unidata.ucar.edu, and:
> 
> exec    "/home/ldm/decoders/xcd_run MONITOR"
> exec    "/home/ldm/bin/pqact -f ANY /home/ldm/etc/pqact.conf"
> exec    "/home/ldm/bin/pqact -f ANY /home/ldm/etc/pqact.gempak"
> #exec   "/home/ldm/bin/pqact -f WMO /home/ldm/etc/pqact.siu.conf"
> exec    "/home/ldm/bin/pqsurf -d /home/ldm/ -p ^S([AP]|US..)"
> exec    "rtstats -f ANY-WSI -h rtstats.unidata.ucar.edu"
> 
> All CRAFT stuff is currently inside my pqact.gempak file. I don't see why
> invoking 4 different ones like you did above helps.

Having multiple pqact invocations will allow your machine to process the
all of the data you want to process as fast as possible.  When pqact.conf
files have a lot of actions, especially a lot of FILE actions, processing
data out of the LDM queue can get bogged down and make it look like all
data was not received.  As I noted in my previous email, we are processing
all of the data available in the IDD on a dual Athlon 2400+ machine, and
its load average is something like 2-5.
 
> Sounds like it works the disk drive even harder, or just the same.

Without looking at your setup, I am at a loss to explain the load average
of 7 when you are trying to process (FILE) all of the Level II data.
At worst, your machine should be impacted by I/O wait, not CPU use.

> So, after my load average fiasco, instead of requesting ".*", I am doing
> "KLOT|KMKX..." etc etc for like 8 sites. My pqact entry hasn't changed.

OK.

> And any idea when radar mosaicking in McIDAS with level 2,

NSSL (Norman, OK) has been working on (3D) compositing the Level II data.  Our
intention is to test the data for usability and eventually make it available
through the IDD.  I want to make you aware of a couple of things in this
effort:

- we have been waiting for a full volume composite for a number of months
  (we are currently receiving tiled composites)

- this is a HIGH volume feed

> and level 2 single site display is coming? Timetable/goal?

This has not a high priority item for McIDAS development since the display
model for the current implementation of McIDAS does not include 3D volumes.
The next version of McIDAS, McIDAS-V, is to be based on VisAD and the Unidata
IDV.  Like the IDV, that version will contain 4D display capabilities (3D
plus time, of course).  If you are itching to play with the Level II data
in a 3D world, you should give the IDV a whirl!

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: LNL-619813
Department: Support IDD
Priority: Normal
Status: Closed