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

20041203: LDM for the SURA SCOOP project

>From:  Gerry Creager N5JXS <address@hidden>
>Organization:  Texas A&M University -- AATLT
>Keywords:  200412031403.iB3E3olI004347 LDM feedtypes

Hi Gerry,

>What's involved in creating a new product feed name on IDD?

There is a limited set of product names in the current implementation
of the LDM (31).  Given this, you will have to use an existing feed
type for the data movement in SCOOP.  I got a call from an old friend
at UAH who is involved in SURA SCOOP asking the same questions (Matt
Smith), and I gave him the same basic explanation (although more
involved).

>In SCOOP 
>(SURA Coastal Ocean Observing and Prediction) we (TAMU, UA Huntsville) 
>are interested in creating a new transport system to augment the current 
>methods in use (OPeNDAP browsing/selection).  We'd like to start a 
>product stream with selected model products, land and sea surface obs, 
>some satellite/RS data.
>
>What will we need to do to keep this legal?

I would suggest using either the EXP or SPARE feed types.  As I told
Matt, if the distribution topology is isolated -- meaning that the
participants in the SCOOP IDD do not interact with participants
in the Unidata North American IDD -- you can use any feed type that
you like.  If the two networks will "interact" -- and Matt thinks that
they will -- then you can still use whatever feed type(s) you like,
but you should try to keep apples with apples (e.g., NEXRAD Level III
data in the NNEXRAD stream) and oranges with oranges (e.g., model data
with something like HDS).  Just remember that if you do mix, your
ingest statistics will not match with those in the Unidata North
American IDD (i.e., the volume in your NNEXRAD feed might be more than
the full feed if you are adding products in addition to the full
NNEXRAD IDD feed).

>I'm thining we can start using the EXPER tag, but would it be possible 
>to engender a SCOOP tag?

EXP, yes.  To name a feed SCOOP, you would have to edit LDM C header
files and rebuild.  The textual feed type name is a representation of
a bit in 32-bit word.  The LDM source code allows for multiple textual
names to be identified as meaning the same bit pattern, but it only
allows for a single primary name.  The primary name is the one that
will be listed, for example, when you run notifyme for a particular
feed type.  The name listed comes from _your_ machine's LDM build,
not from the upstream's build.

Please let me know if you have further questions about this.

>Thanks, Gerry

Cheers,

Tom
--
+-----------------------------------------------------------------------------+
* Tom Yoksas                                             UCAR Unidata Program *
* (303) 497-8642 (last resort)                                  P.O. Box 3000 *
* address@hidden                                   Boulder, CO 80307 *
* Unidata WWW Service                             http://www.unidata.ucar.edu/*
+-----------------------------------------------------------------------------+