[AWIPS #KCP-365241]: AWIPS2 system hardware specs and software availability

Hi Iain,

re:
> We met briefly at last year's AMS annual meeting in New Orleans, and
> I wanted to stop by to see you at the Unidata exhibit booth at the AMS
> meeting in Austin, but each time you were busy, so sorry about that.

It is too bad about missing each other at AMS!

re:
> What I wanted to ask was as follows;
> 
> 1. Will Pelmorex be able to have access to the AWIPS2 software by virtue
> of our long-standing relationship with Unidata? (as we have done in
> the past with GEMPAK and NAWIPS)

Yes.

re:
> 2. Can you give me an idea of the hardware requirements for running
> AWIPS2?

This is a LOT harder to do as we are still trying to determine the
minimum machine on which AWIPS-II EDEX can run well (EDEX is the
data ingest/decoding/serving component of AWIPS-II).  Currently,
EDEX is hamstrung by the requirement to run on an old version of
32-bit Linux, RedHat Enterprise 5.6 (officially, we are running
on 5.8).  We are eagerly anticipating a release that supports
running on 64-bit Linux, and equally hoping for a port to a
current version of Linux (e.g., RedHat Enterprise 6.3 x86_64).
We have been informed on numerous occasions that this is the
direction that development is heading, but we have been given no
hard estimate for when that release will be made (even in beta).

Suffice it to say that we are currently testing on a robust
machine that has:

Dual quad-core CPUs
24 GB RAM
1 TB HDD

Even with this configuration, we (Michael James) have encountered
bottlenecks in EDEX processing... one example is that this machine
can _not_ keep up with the full complement of NEXRAD Level III
products currently being sent in NOAAPort.  A "solution" has been
floated (more threads dedicated to NEXRAD Level III processing),
but an interface for doing this has not been exposed to the
end-user; one would have to have a full AWIPS-II development
environment to be able to build a distribution that supported
more than one thread for Level III processing, and we are not there
yet.

As we (again, Michael) work through testing we will have a much
better idea of the minimum (set of) hardware that can be used
for EDEX.  We will, of course, be letting users know what this
spec is as soon as we are able.

re:
> Many thanks in advance....

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: KCP-365241
Department: Support AWIPS
Priority: Normal
Status: Closed