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

Re: CONDUIT grids getting later and later again

CONDUIT folks,

I think Pete Pokrandt has summed up all of my (U.Wash Atm. Sci.)
votes, concerns, and questions quite well.  Ditto.

-- 
David Ovens              e-mail: address@hidden
Research Meteorologist    phone: (206) 685-8108
Dept of Atm. Sciences      plan: Real-time MM5 forecasting for the
Box 351640                        Pacific Northwest
University of Washington          http://www.atmos.washington.edu/mm5rt
Seattle, WA  98195               Weather Graphics and Loops
                                  http://www.atmos.washington.edu/~ovens/loops


On Fri, Feb 04, 2005 at 11:12:41AM -0600, Pete Pokrandt wrote:
> 
> In a previous message to me, you wrote: 
> 
>  >Pete,
>  >
>  >There are several options that feed back from the users would be helpful
>  >in moving forward:
> 
> Steve,
> 
> Thanks for getting back to me on this so quickly. My thoughts below:
> 
>  >
>  >One pending action item from Last June was to remove the ETA 
>  >awip20 (grid 215) files when ETA218 was available. Does anyone 
>  >rely on these files from CONDUIT?  Would the ETA218 grids in 
>  >NOAAPORT (GRIB2 format in DVB-S) eliminate the need for 
>  >grid 215 in CONDUIT?
> 
> We (UWisc-AOS) could definitely live w/o the 215 grids. We're currently 
> not using them at all. That doesn't really reduce the data stream
> all that much though, compared to the increase from the 0.5 deg GFS,
> does it?
> 
>  >
>  >As an alternative to removing the 0.5 degree GFS, would it be useful 
>  >to keep the highest resolution grids available for f000-f084 and 
>  >remove the 1 degree grids for the period of time where the two sets 
>  >overlap?
> 
> This is a good idea, however I suspect many of us are using software
> that does not yet read the GRIB2 format files. How easy or hard would 
> it be to come up with a conversion utility that would read the 0.5 deg
> grib2 data and convert into either 0.5 deg GRIB (with similar format,
> parameters that we currently have) or to average out to a 1 deg format
> in the same format as the current 1 deg files? Is this even feasible?
> 
>  >
>  >There are a number of data sets which users have expressed interest 
>  >in adding, though clearly it is not advisable to be adding with the 
>  >current limitations.
> 
> I agree, adding anything at this point would only make things worse.
> 
> Would having multipile ingest boxes, each doing different parts of
> the conduit feed make any difference, or would that just pass the
> delay and backlog downstream to the top level relays?
> 
> How about another "experimental" conduit feed, or one for new, or
> GRIB2 formatted data that the new data could be tested on without
> any expectation of regular timely delivery? I guess CONDUIT does
> not 'guarantee' timely delivery, but it has been so reliable over
> the past couple of years that many of us have come to depend on
> the products in it, and pretty much expect it to be reliable.
> 
> It seems to me that the problem we are seeing, or trying to get 
> around, is that there are short periods where LOTS of data becomes 
> available, and while we'd all like to get the data as soon as it is 
> available, the current latency of the network and/or ldm is just not 
> capable of handling the large bursts in a reliable and predictable 
> manner.
> 
> My rambling thoughts..  Personally, I'd still vote to remove the 0.5 deg 
> gfs data from the CONDUIT stream for now..
> 
> Pete
> 
>  >
>  >Steve Chiswell
>  >Unidata User Support
>  >
>  >On Fri, 4 Feb 2005, Pete Pokrandt wrote:
>  >
>  >>
>  >> All,
>  >>
>  >> 06 UTC GFS didn't start coming in here until after 12 UTC this AM.
>  >> It's just now finishing up at 14:31 UTC.
>  >>
>  >> I'd like to second Robert's suggestion from a few weeks ago that
>  >> we back out of the 0.5 deg GFS data until we can resolve these
>  >> speed issues.
>  >>
>  >> I haven't had time to even look at the 0.5 GFS data yet; is anyone
>  >> using it?
>  >>
>  >> Pete
>  >>
>  >> --
>  >> +>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>+<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<+
>  >> ^ Pete Pokrandt                    V 1447  AOSS Bldg  1225 W Dayton St^
>  >> ^ Systems Programmer               V Madison,         WI     53706    ^
>  >> ^                                  V      address@hidden       ^
>  >> ^ Dept of Atmos & Oceanic Sciences V (608) 262-3086 (Phone/voicemail) ^
>  >> ^ University of Wisconsin-Madison  V       262-0166 (Fax)             ^
>  >> +<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<+>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>+
>  >>
> 
> 
> --
> +>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>+<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<+
> ^ Pete Pokrandt                    V 1447  AOSS Bldg  1225 W Dayton St^
> ^ Systems Programmer               V Madison,         WI     53706    ^
> ^                                  V      address@hidden       ^
> ^ Dept of Atmos & Oceanic Sciences V (608) 262-3086 (Phone/voicemail) ^
> ^ University of Wisconsin-Madison  V       262-0166 (Fax)             ^
> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<+>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>+