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

Re: Recent Version of LDM Required to Receive NEXRAD Data

Gregory Grosshans wrote:
> 
> Presently the SPC is using modified acqserver code, originally from FSL,
> to inject 2-4 NOAAPORT channels into LDM 5.1.2.  All products are being
> inserted in the queue with the same feedtype (e.g. WMO).  The NWSTG
> products are modified to look like a FOS feed.  The most recent version
> of LDM allows numerous feed types for the various data sets flowing
> through NOAAPORT.    The SPC uses LDM to receive and move data around
> in-house.  We don't feed any data sets across our WAN, and if we do, the
> downstream user would likely request  a subset  with the 'request' line.
> Can you tell me is there any advantage to LDM performance if one actually
> breaks up the NOAAPORT stream and assigns various feedtypes to different
> product categories?
> 
> Thanks,
> Gregg

Hi Gregg,

I can think of only one significant way that multiple feed
types might improve performance.  I suspect there would be a
performance improvement in matching patterns to product IDs,
i.e., I assume that the LDM would first look at a product's
feedtype and rule out products with non-matching feed types
before attempting to match a pattern against the ID.  Since
pattern matching can be costly performance-wise, eliminating
a significant number of products from that test might make a
difference.

Anne
-- 
***************************************************
Anne Wilson                     UCAR Unidata Program            
address@hidden                 P.O. Box 3000
                                  Boulder, CO  80307
----------------------------------------------------
Unidata WWW server       http://www.unidata.ucar.edu/
****************************************************