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

Re: 20051011:QPE "mosaic"



Hi Steve,

We are currently discussing the pros and cons of both scenarios
(reformatting and stitching locally or delivering the full mosaic)...and
how it may work within the LEAD and THREDDS environments...

Can you open up the feed of the tiles to us so we can have better (more
timely) access to the feed for experimentation?

Again it would be best if any machine from the unidata.ucar.edu domain
could access the data stream as we hope to be able to be dynamic in which
machine is ingesting the data.. we will NOT pull from you more than
one feed at a time.

What machines are currently relaying the QPE data? (where should we point
our requests).

Cheers,

Jeff
---------------------------------------------------------------------
Jeff Weber                                    address@hidden        :
Unidata Program Center                        PH:303-497-8676        :
University Corp for Atmospheric Research      3300 Mitchell Ln       :
http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000  :
---------------------------------------------------------------------

On Mon, 24 Oct 2005, steven vasiloff wrote:

> Jeff,
>
> I was wondering if you all could run the reformatting and the stitching
> programs there...
>
> Or, would it be better to send you single CONUS files?
>
> Steve
> On Oct 24, 2005, at 10:39 AM, Jeff Weber wrote:
>
> > Hi Steve,
> >
> > Thanks for the update, let us know if we can help re the netCDF file
> > format...
> >
> > Cheers,
> >
> > Jeff
> > ---------------------------------------------------------------------
> > Jeff Weber                                    address@hidden        :
> > Unidata Program Center                        PH:303-497-8676        :
> > University Corp for Atmospheric Research      3300 Mitchell Ln       :
> > http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000  :
> > ---------------------------------------------------------------------
> >
> > On Mon, 24 Oct 2005, steven vasiloff wrote:
> >
> >> Jeff,
> >>
> >> Wenwu wrote me that the sizes of gunzipped mosaic2d and mosaic 3d
> >> files
> >> for each tiles are:
> >>   tile                            mosaic2d              mosaic3d
> >>   1                               60M                       54M
> >>   2                                60M                      54M
> >>   3                               30M                       27M
> >>   4                                60M                      54M
> >>   5                               80M                       72M
> >>   6                                80M                      72M
> >>   7                               40M                       36M
> >>   8                                80M                      72M
> >>
> >> We have some work to do to get our program running that stitches the
> >> tiles into a CONUS file.  2D will be ~500 MB and 3D ~400 MB.  We may
> >> need to provide a format converter since we output a modified netCDF.
> >>
> >> Also, I we are fine tuning our radar QC to ensure a quality product.
> >> Once we finalize these things we will start shipping you all the data.
> >> The LEAD connection will be important for us and I am pleased that you
> >> are in a position to help us with it.
> >>
> >> I will review the status of these issues with Jian, Ken and Wenwu and
> >> get you a time estimate ASAP.
> >>
> >> Steve
> >> On Oct 20, 2005, at 11:11 AM, Jeff Weber wrote:
> >>
> >>> Hi Steve,
> >>>
> >>> Any word on allowing us to feed from you the 2d and 3d gridded
> >>> mosaic?
> >>>
> >>> Thanks,
> >>>
> >>> Jeff
> >>> ---------------------------------------------------------------------
> >>> Jeff Weber                                    address@hidden
> >>> :
> >>> Unidata Program Center                        PH:303-497-8676
> >>> :
> >>> University Corp for Atmospheric Research      3300 Mitchell Ln
> >>> :
> >>> http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000
> >>> :
> >>> ---------------------------------------------------------------------
> >>>
> >>> On Tue, 11 Oct 2005, Jeff Weber wrote:
> >>>
> >>>> Hi Steve,
> >>>>
> >>>> We have not found the "upper limit" yet, TCP/IP breaks everything up
> >>>> into
> >>>> chunks anyway...we ingest imagery that is > 25 MB per file, not to
> >>>> mention
> >>>> hi-res model output :)
> >>>>
> >>>> We are primarily interested in the 2d and 3d gridded mosaic.(in
> >>>> netCDF)
> >>>>
> >>>> Are they larger than 4-500MB? Would not want to crater the queue,
> >>>> that is
> >>>> the main concern....our queue's range from 8-12GB
> >>>>
> >>>> The allow line for us is:
> >>>>
> >>>> allow  EXP  \.unidata\.ucar\.edu$
> >>>>
> >>>> This will allow us to ingest on different machines for testing, we
> >>>> will
> >>>> only draw one feed at a time.
> >>>>
> >>>> ...and what machine would we point this request at?
> >>>>
> >>>> Feel free to pass any compatibility questions re THREDDS or LEAD by
> >>>> me
> >>>> (us) at anytime...
> >>>>
> >>>> Thanks for working with us...
> >>>>
> >>>> Cheers,
> >>>>
> >>>> Jeff
> >>>> --------------------------------------------------------------------
> >>>> -
> >>>> Jeff Weber                                    address@hidden
> >>>>  :
> >>>> Unidata Program Center                        PH:303-497-8676
> >>>>  :
> >>>> University Corp for Atmospheric Research      3300 Mitchell Ln
> >>>>  :
> >>>> http://www.unidata.ucar.edu/staff/jweber      Boulder,Co 80307-3000
> >>>>  :
> >>>> --------------------------------------------------------------------
> >>>> -
> >>>>
> >>>> On Tue, 11 Oct 2005, Steven Vasiloff wrote:
> >>>>
> >>>>> Jeff,
> >>>>>
> >>>>> 1)  I will check on the file sizes of CONUS products.  What is a
> >>>>> max
> >>>>> size to efficiently transfer?
> >>>>>
> >>>>> 2)  THis is definitely possible and I can talk to our IT guy about
> >>>>> setting this up.  The solution somewhat depends on the answer to
> >>>>> 1).
> >>>>>
> >>>>> Also, we are working with Kelvin to leverage NMQ/JADE (a web portal
> >>>>> for
> >>>>> collaborative research onliine) with the LEAD project.  We will be
> >>>>> developing XML SOAs and would like to be compatible with thredds
> >>>>> etc.
> >>>>> More on this later.
> >>>>>
> >>>>> Steve
> >>>>> On Oct 11, 2005, at 1:48 PM, Jeff Weber wrote:
> >>>>>
> >>>>>> Hi Ken and Steve,
> >>>>>>
> >>>>>> I am writing you in regards to the QPE national mosaic product.
> >>>>>>
> >>>>>> I have two questions:
> >>>>>>
> >>>>>> 1) Do you plan to migrate away from the "tiled" products and
> >>>>>> deliver a
> >>>>>>    single file for the CONUS?..if so, when do you plan to do this?
> >>>>>>
> >>>>>> 2) Can we obtain a direct feed to Unidata via IDD/LDM rather than
> >>>>>> feeding
> >>>>>>    from UCAR/RAP?
> >>>>>>
> >>>>>>
> >>>>>> Cheers,
> >>>>>>
> >>>>>> Jeff
> >>>>>> ------------------------------------------------------------------
> >>>>>> --
> >>>>>> -
> >>>>>> Jeff Weber                                    address@hidden
> >>>>>>  :
> >>>>>> Unidata Program Center                        PH:303-497-8676
> >>>>>>  :
> >>>>>> University Corp for Atmospheric Research      3300 Mitchell Ln
> >>>>>>  :
> >>>>>> http://www.unidata.ucar.edu/staff/jweber      Boulder,Co
> >>>>>> 80307-3000
> >>>>>>  :
> >>>>>> ------------------------------------------------------------------
> >>>>>> --
> >>>>>> -
> >>>>>
> >>>>
> >>
> >>
>
>