Re: 20040419:Level II Feeds

Hi Matt,

Comments in text:

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, 19 Apr 2004, matthew huber wrote:

>
> On Monday, April 19, 2004, at 03:24 PM, Jeff Weber wrote:
>
> > Hi Matt,
> >
> > We expect the 2nd tier sites to have I2 connections as our focus is the
> > .edu domain.
> >
>     I believe a number of potential meteorology colleges and
> universities do not have I2 connectivity (or it isn't
> very good).


Then they would not be candidates for a second tier connection.



> > Not sure what hardware you are speaking of that is (could be) delayed
> > by
> > this decision...
> >
> > The decision is based upon connectivity and reliability of
> > administration
> > as well as logical geographical considerations.

> depending on these reliability and connectivity concerns we will have
> to 'cache' differing amounts of
> data and at 80Mb/s that adds up pretty fast in terms of cost, also it
> has some impact on when we upgrade from
> 32bit to 64bit architectures for the server (for memory footprint
> reasons).



Most sites are running a 2Gb queue, with large file support we can
generate queues in the range of 10-20 GB's if needed.

32 bit architecture should be adequate for these needs.

Your archive needs may require additional disk space, but that is your
call.


>
>
> >
> > Which network topology are you speaking of:
> >
> >> where in the network topology?
> >
> > The level II data distribution will have its own topology.
> >
> > Who do you desire to feed...?
> There are several topologies, I2, the level II data on top of that, and
> the ETF.  Ideally, we'd like at least some of our downstream peers to
> be on ETF, which may have some very nice performance attributes
> provided we can work to get the streaming data agreed upon by various
> national committees as a valid part of ETF.
>


Perhaps we are confusing the distribution of level II data to the arena of
grid computing..

Not quite sure...perhaps a phone call is desirable to clear up any
confusion.


PH:303-497-8676


> a partial potential list includes
>       Argonne/U of Chicago
>       NCSA/UIUC
>       Indiana University
>       Oak Ridge National Lab/Georgia Tech
>            UCSD
>       Caltech
>
> -m
>
> >
> > Thank you,
> >
> > 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, 19 Apr 2004, Unidata Support wrote:
> >
> >>
> >> ------- Forwarded Message
> >>
> >>> To: Linda Miller <address@hidden>
> >>> cc: <address@hidden>
> >>> From: Matthew Huber <address@hidden>
> >>> Subject: Re: Unidata
> >>> Organization: UCAR/Unidata
> >>> Keywords: 200404191941.i3JJfuCT015056
> >>
> >> Hi,
> >>    It's been a while since I put in my first request, so I'll request it
> >> again.  What is the process for determining the second tier sites.  We
> >> have a number of hardware purchases that are on hold until we can
> >> determine who our downstream peers are....Internet2 sites, ETF sites,
> >> where in the network topology?  We've got preferences for who we'd
> >> like
> >> to distribute to as well.
> >> thanks,
> >> -matthew
> >> On Apr 19, 2004, at 1:50 PM, Linda Miller wrote:
> >>
> >>> All,
> >>>
> >>> Unidata is pleased to offer assistance to you concerning topology
> >>> decisions for relay site distribution of Level II data.  Unidata
> >>> Support will also provide recommendations for second level sites and
> >>> beyond.
> >>>
> >>> Unidata's Internet Data Distribution (IDD) has been successfully used
> >>> for data distribution since 1995, so we have a great deal of
> >>> experience with coordination and maximizing data flows for data
> >>> distribution.
> >>>
> >>> Unidata support will respond promptly concerning your inquiries,
> >>> including support and statistics that will facilitate the process.
> >>> The documentation available from Unidata should be a useful tool for
> >>> you as you ramp up.  Support will make recommendations on support
> >>> email lists, Web documentation, etc., that will be useful to you.
> >>>
> >>> As mentioned during the recent conference call, send your email to:
> >>>
> >>> <address@hidden>
> >>>
> >>> Best of luck to all of you as you embark on this new endeavor!
> >>>
> >>>
> >>>
> >>> Linda Miller - address@hidden
> >>> External Liaison, Unidata
> >>> University Corporation for Atmospheric Research
> >>> P.O. Box 3000
> >>> Boulder, CO 80307-3000
> >>> 303-497-8646 fax: 303-497-8690
> >>>
> >>
> >> --
> >> **********************************************************************
> >> Unidata User Support                                    UCAR Unidata
> >> (303)497-8643                                                  P.O.
> >> address@hidden                                   Boulder,
> >> ----------------------------------------------------------------------
> >> Unidata WWW Service
> >> ----------------------------------------------------------------------
> >> NOTE: All email exchanges with Unidata User Support are recorded in
> >> the
> >> Unidata inquiry tracking system and then made publically available
> >> through the web.  If you do not want to have your interactions made
> >> available in this way, you must let us know in each email you send to
> >> us.
> >>
> >> ------- End of Forwarded Message
> >>
> >
>