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

[McIDAS #GUB-772131]: McIDAS - Cannot contact ADDE East and West Servers



Hi Fred,

> Since the last problem of Mcidas data access last month, I have been
> concerned about the timeliness of the access to GOES-east and west data.  It
> is often 30 minutes late, and is causing me problems when I try to do our
> global mosaic in that Meteosat is frequently 30 minutes off from GOES-east.
> Meteosat is available at the end of each scan sequence, but GOES is not.

> So my questions for data access:  Is the transfer done only after data is
> complete, such as with FTP or LDM?

I am not sure of the exact logic that is used in determining when data can
be moved from the SDI server to unidata2.ssec.wisc.edu as I did not create
it.

> What time delay is assumed for
> determining the data is complete, or is this determined dynamically by
> looking at the file close?

I don't know, but I will ask.

> Why aren't ADDE mcidas commands being utilized
> for these transfers?

ADDE transfers would convert the original data, which are in a raw scan
format that is very compact, into McIDAS AREA files which would occupy
at least 100 times (and perhaps 1000 times) more disk space for the same
time record of images. unidata2 does not have the disk space available for
this.

> What is the best strategy to get timely data?  Should
> I request data at standard times; or should I check every minute or so to
> see if new data is available; or is there a better way of knowing when to
> ask for data?  "  Dee has not yet responded to my latest e-mail.

I believe that the arrival time of the data on unidata2 is fairly consistent.
As soon as some network firewall issues are resolved, we will be able to
monitor the receipt remotely from our machines and determine the schedule
as best we can.  Once the schedule is known, you can adjust the kickoff of
your processing to better insure use of the most recent data available.
Please do not run a job every minute as this would not make good use of
unidata2 resources.

> So my latest thought is maybe I am complaining to the wrong person.

You were complaining to the wrong group.  My belief is that part of the
arrangement we have with SSEC to provide the full GOES-East/West data is
to provide a buffer for end-user inquiries.

> Is the
> LDM data transfer logic of the GOES data from the SSEC server to the Unidata
> server under your control or SSEC's control?

The process on the SDI server was done by SSEC, and we do not have login access
to that server (or its backup).

> Can anything be done to improve
> the timeliness of the data for all the different scan modes?

I will ask.

> I favor using
> ADDE data transfers for Mcidas data rather than LDM because it would allow
> for data scan following by the users, thus allowing truly real time data for
> all users.

I do not favor using ADDE transfers from the SDI machines to unidata2.  The
principal reason is the amount of disk space that would be needed to store the
imagery in McIDAS AREA format.

Currently you have a second option for GOES-West data:  use the ADDE server
on the GOES-West SDI machine that we maintain here in UCAR:

group server
-----+-------------------------
WEST  goeswest.unidata.ucar.edu

Concurrent with my release of Unidata McIDAS v2006, I updated:

Publically accessible McIDAS ADDE servers
http://www.unidata.ucar.edu/software/mcidas/adde_servers.html

to include the goeswest.unidata.ucar.edu machine.  Your IMGCOPY access of
data off of this machine will track images as they are received, so the
GOES-West component of your composites will be as up to date as possible.

We are in the process of working with the RAL group in NCAR to get
access to their GOES-East and GOES-West TeraScan (tm) ingest streams.  Our
intention is to provide SDI ingest and ADDE access to the GOES-East data like
we have done with the GOES-West data on goeswest.unidata.ucar.edu.  We
are also talking with the RAL folks about installation of a third dish
so we can both ingest GOES-10 South American data whenever it gets on station
at 60 W.

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: GUB-772131
Department: Support McIDAS
Priority: Normal
Status: Closed