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

Re: Plan for testing NEXRAD distribution (fwd)


===============================================================================
Robb Kambic                                Unidata Program Center
Software Engineer III                      Univ. Corp for Atmospheric Research
address@hidden             WWW: http://www.unidata.ucar.edu/
===============================================================================

---------- Forwarded message ----------
Date: Mon, 13 Nov 2000 13:12:40 -0700
From: Steve Chiswell <address@hidden>
To: Anne Wilson <address@hidden>
Subject: Re: Plan for testing NEXRAD distribution



Anne,

Please use motherlode or thelma as the source- not jackie, since
the SSEC ingestor source is desi and not jackie. Jackie is my ingestor,
which is something different.

We can configure motherlode or thelma to request the NEXRAD feed from
desi. The pqing change I implemented on desi does not differentiate
between encrypted and unencrypted. There is only 1 site (KARX) currently
broadcasting unencrypted data (to provide a test for sites to prepare for
the unencryption of all data after Jan 1, 2001. So, you will want to
send the entire feed.

Chiz



On Mon, 13 Nov 2000, Anne Wilson wrote:

> Hi Russ, Chiz, and Robb,
> 
> I have developed a plan for testing the distribution of the NEXRAD data and 
> am interested in any thoughts about it.
> 
> With Jeff's help, we picked a path starting with Gilbert's site (squall) 
> followed by Clint's site (chinook).  I figured at least these two would be 
> strongly supportive of the test.
> 
> chinook can be followed by either cirrus.smsu.edu or whistler.creighton.edu.  
> Both of these leaf nodes are reporting statistics on the FOS routing page, so 
> that might be informative.  With respect to FOS routing, cirrus is currently 
> reporting a delay of 21 minutes while whistler is reporting only 5 minutes, 
> so the contrast between the two may also be informative. 
> 
> This gives us a three node path (not counting motherlode).  There are other 
> paths that have more nodes, but this one has fairly 'average' bandwidth, 
> according to Jeff, as opposed to other first tier sites that have better 
> connectivity. 
> 
> My thought is to have each site, starting with squall, request all the data 
> first without filing it, followed by a test where they do file the data. 
> (This part could be broken down even further based on the filing scheme.)  
> Once a site is filing the data successfully, subsequent sites can be added.
> 
> From the email, I see that we now have unencrypted, compressed NEXRAD data.  
> Is that the one I should be using, being what the final products will be 
> like? or are we not yet supposed to distribute that?  Truthfully, I don't 
> understand why it has come about that this data is now available.
> 
> If that is the data to use in the test, then I would have to ask these sites 
> to feed from jackie since that's where they can be differentiated based on 
> the product ID.  But, is feeding from jackie really representative?
> 
> Thoughts?  Comments?
> 
> Anne
>