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

[Support #TOL-308327]: statistics

Hi Yoshihiro,

> I installed the new version on both computer's (atm77 and
> pp42) and I am requesting 00 & 12 UTC conduit data on
> atm77 and 06 18 UTC datga on pp42. The results loos very
> good now.

Yes, the results look very good.  Just so you know: you should
have no trouble ingesting all NCEP GFS data on atm77 (0, 6, 12,
and 18 UTC) if you would rather not use two machines.

> By the way since cptec/inpe (moingobe) was having similar
> latency problem they also made the partition request and
> now seems fine too.

I know.  Right after the splitting experiment at your site
showed progress I contacted Waldenio and asked him to split
the CPTEC CONDUIT request into fifths and then tenths.  The
results for the CPTEC feed is that 10 requests kept the
latencies almost as low as could be hoped.  Reverting back
to a 5-way split is working, but peak latencies are on the
order of 300 -400 seconds.  The next change I will ask Waldenio
to make is to keep the 5-way split, but to mimic the setup
you are now running: redundant 5-way splits, one to idd.cise-nsf.gov
and the other to idd.unidata.ucar.edu.  In fact, I will be
sending the request to Waldenio as soon as I finish this

I am glad to see that your CONDUIT data ingest is now working
the way that it should!

> Best regards,
> yyamazaki


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: TOL-308327
Department: Support IDD
Priority: Normal
Status: Closed