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

Re: New NIDS products question...

Gregory Grosshans wrote:
> 
> Anne,
> 
> I was curious if you forsee some required change to the ingest software 
> Unidata uses
> to ingest the additional radar data?  We are currently using a modified 
> version of
> the FSL acqserver ingesting the NWSTG, GOES EAST & WEST into LDM (5.1.2).  If 
> the
> data isn't a NEXRAD product its injected into LDM as a 'WMO' feedtype.  If 
> its radar
> data I inject it as NNEXRAD.  This same LDM also ingests an X.25 line with the
> equivalent of FOS (PPS/DDS), most of the products are dup'ed out as a result 
> of the
> NOAAPORT feed,  and a Unisys radar feed containing individual radar data and 
> mosaic
> data inserted as feedtype 'WSI.'  The queue on this system is 750 MB.
> 
> My initial thoughts are the data will come in without any problem.  At worst I
> perhaps will need to increase the queue size.
> 
> Thanks,
> Gregg
> 
> ldmalice@alice> pqmon -q  /ops_local/ldmalice/data/ldm.pq
> Jul 17 14:05:59 pqmon: Starting Up (10732)
> Jul 17 14:05:59 pqmon: nprods nfree  nempty      nbytes  maxprods  maxfree
> minempty    maxext  age
> Jul 17 14:05:59 pqmon:  31116     6  151983   664123120     70379       43
> 112725  20957592 4124
> Jul 17 14:05:59 pqmon: Exiting
> ldmalice@alice>
> 

Hi Gregg,

Yes, we will modify the ingest software to accomodate the new products. 
I've heard of the acqserver, but otherwise I don't know anything about
it.  We'll be modifying the pqing program in a manner similar to what it
does now.  Currently it is invoked specifying the WMO feedtype, but
SDUS5* products are relabeled as NEXRAD.  (Sounds like what you do.) 
We'll have to add logic to test for these additional products.

From your pqmon output, it looks like your queue is pretty well tuned
for the current conditions, at least for that moment in time - you've
got a little over an hour's worth of data and queue usage is not quite
within 100Mb of the limit.  (These values may vary depending on what
time you ran pqmon.)  You may be able to squeeze in the additional data,
but it wouldn't hurt to increase the size a bit.  I don't think we'll be
able to get an estimate on the volume of additional data ahead of time,
so everyone will probably wind up retuning their queues once the data is
flowing.  We'll certainly convey whatever we find out.

So, you're getting *all* of NWSTG, GOES EAST and WEST in your 750Mb
queue?  That's an interesting data point for me.

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