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

Re: New Client Reply - [Support !SBB-325304]: Re: 20110712: CONDUIT request -- fire weather grids

OK, I thought perhaps that was the case, but wanted to check anyway. The Policy Committee discussion came from a meeting that took place there last year.
Let's stick to the Rapid Refresh, if everybody agrees. Tom needs to check in to let us know if that is doable with today's bandwidth, decoders, etc. 

Linda

On 8/31/2011 1:34 PM, Rebecca Cosgrove wrote:

Hi Linda.
Now hold on a minute...  we're starting to talk about different things :)
When I say Rapid Refresh, I mean the one that'll be put into operations here at NCO. Developed by GSD and implemented here. It's the RUC replacement, so 13km is the best it can do.
The HRRR right now is only available from GSD. There is some talk of us hosting GSD's experimental HRRR dataset, but that's in the very infant stages. And someone had asked about the FIM in an unrelated conversation and I was told that was way, way off from an NCEP perspective at least. Now if we start hosting the HRRR, who knows what that opens the door to.
So what I was offering up was just the RR, which no High-Resolution. So the version of the Rapid Refresh that's on its way to being implemented hopefully in late Fall. But we could post grids from it a bit sooner than it's go-live date. 

Becky

On 8/31/2011 3:28 PM, Linda Miller wrote:

Hey Becky,
I agree with the suggestion that we go forward with the Rapid Refresh. There were some on the Policy Committee who definitely wanted HRRR. They had suggested making the it available via THREDDS. They were also interested in the FIM. Do you have that, or isn't that ready yet? 

Linda

On 8/31/2011 12:39 PM, Rebecca Cosgrove wrote:

New Client Reply: Re: 20110712: CONDUIT request -- fire weather grids

Tom and Linda,
I think I found the information! I sent you earlier today the blurb EMC 
wrote up about the Fire Weather products.  So that was one request you
had for me.  The other was what the plans were for adding data to
CONDUIT in the near-term.  Here's that part:

  From the meeting in March, I have noted that I had the okay to add
NAM Fireweather grids
NAM 221 grid out to 84 hrs.   This is a 32km North America grid.

Then we polled the community.  The only responses I found in my inbox
were two that came through Kevin T, and they both requested very
different things.  Respondent 1 said, in order:
HIRESW
Rapid Refresh
NAEFS ensemble
GFDL
HWRF
RTOFS Atlantic

Respondent 2 said:
GFS analysis/forecasts
NAM analysis/forecasts
Rapid Refresh analysis/forecasts
RTMA

So the one in common is the Rapid Refresh.  I think we could work out a
parallel feed once the NAM goes in in late September.  I'm not sure
there's a clear enough signal beyond that. I could throw in some NAEFS, but that will come during that peak time we're trying to avoid until the 
upgrade.

What if we say we'll add the Fire weather around 9/20 and the NAM 221
shortly after that.  Then we'll hold anything else new except maybe an
early look at the Rapid Refresh till after the upgrade of my boxes in
Nov/Dec.  That way I can come out in late October and we can talk about
what else we want to do post-upgrade.

Do you think we could also pull out the 90 km NAM fields?  My
presentation says they're getting 40km and 90km and a limited set of
surface fields from the 12.  Hopefully they could live without the 90?

Becky
On 8/26/2011 10:26 AM, Unidata CONDUIT Support wrote:

Hi Becky,

re:

There is a lot of good information in your email.  Thanks for the
careful analysis.

I try :-)

re:

While we had ldm2 pulled out of the pool I was going
to move on to some of the other additions I've committed to making, but 
I'll have to review the list in light of some of this information.
Can you give me an idea of the other things you were considering adding? 

re:

As far as where we go from here for the fire weather, I think from a
technical standpoint, we would be ready to implement this into
operations.

Excellent!

re:

Technically the fire weather products are still
"experimental" or some such label until September 20th, but we had said 
we'd get them out as soon as we were ready.
Experimental vs operational is not a big issue for the Unidata community. 
In fact, one of the standing requests from a long-time CONDUIT user is
for a parallel version of RTMA grids.

re:

So I think the only
remaining issue is your point about the potential impact to the user
community of the added volume.  How do you want to address that
concern?

I will get together with Linda to discuss this on Monday (she is out
today).  I think that we (Unidata) need to send a note to the CONDUIT
community (via the address@hidden and address@hidden 
email lists) explaining a proposal to add the fire weather products to
the datastream.  This email needs to address:

- why the products are useful/important

    Can you write a little blurb that covers this?

- what the impact will be for those who want to continue to drink the
    entire firehose of data

- what steps can be taken to mitigate the local impact of the addition
of the data (e.g., noting that a site can request everything except 
    the fire weather products, etc.)

- a target timeframe for the addition of the products if there is
    general approval in the community

- some idea of where CONDUIT is heading

    What I have in mind here is information on new, beefier top level
    relay machines that will eventually be installed at the WOCs;
    perhaps a list of additional products that will be added at
    some time; a piece that looks down the road at things like
    0.25 degree global GFS, etc.

re:

Do we need to poll them somehow, or just send a notification
saying this is going to happen and will be a big volume, and here's how 
you opt out?

I think we need buy-in.  Given that sites can chose to not receive the
added data, this shouldn't be has hard as it might seem.
BTW, the other impact will be on the toplevel CONDUIT relay nodes. They will be wanting to relay the data, so I will need to make sure that they understand the impacts. What I have in mind is a "personal" note to each 
toplevel IDD administrator that reviews the sorts of things I sent you
and Justin yesterday.

re:

But as I said, from a technical point, I think we just
have to file paperwork and get approval. In theory, Justin could write 
up something in time for our Tuesday COB deadline which would let us
implement the week of September 5th.

This may be too quick for the university community.  We will know more
when get some feedback from them.  Again, I will meet with Linda on
Monday to discuss the way forward here in Unidata.

re:
So let us know what you think about the customer support part of this. 
Will do.

re:

Thanks!

No worries.

Cheers,

Tom
--
**************************************************************************** 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: SBB-325304
Department: Support CONDUIT
Priority: Normal
Status: Closed




Ticket Details
===================
Ticket ID: SBB-325304
Department: Support CONDUIT
Priority: Normal
Status: Open
Link: https://www.unidata.ucar.edu/esupport/staff/index.php?_m=tickets&_a=viewticket&ticketid=18485 



--
Linda Miller - address@hidden
Community Services, Unidata
University Corporation for Atmospheric Research
P.O. Box 3000
Boulder, CO 80307-3000
303-497-8646 fax: 303-497-8690