[Sis-ams] Draft documents

[Kizzort, Brad]

Scott,

There is a real-time publish subscribe specification that was adopted within OMG
last year.  It was co-written by OIS, RTI, and Thales, so it is probably similar
to the RTPS IEC standard, although I have not looked at the IEC standard.  The
OMG specification is freely available from www.omg.org in the documents section.
Its document ID is: formal/04-12-02.pdf, Data Distribution Service.  RTI and
Thales are both offering or going to offer compliant products and there is an
open source DDS effort underway.  

DDS is currently a model-based specification, but there is a definite interest
in defining wire protocols for interoperability between products much like IIOP
provides for CORBA.  That specification effort has not been kicked off, yet, but
I believe a Request For Proposal is imminent.    

Brad Kizzort

-----Original Message-----
From: sis-ams-bounces at mailman.ccsds.org
[mailto:sis-ams-bounces at mailman.ccsds.org]On Behalf Of Scott Burleigh
Sent: Thursday, March 31, 2005 10:52 AM
To: sis-ams at mailman.ccsds.org
Subject: [Sis-ams] Draft documents


Hi.  I have posted to the sis-ams CWE
(http://public.ccsds.org/sites/cwe/sis-ams/default.aspx) the Requirements list I
sent out last week and also a draft Charter and Resource Plan.  We unfortunately
don't have a lot of time to discuss these things by email between now and the
April meeting, but I think we should still be able to come to some resolution on
progressing to Working Group status by the end of the Athens session.  I've also
posted to the CWE the Agenda for that meeting, which should additionally be
available elsewhere on the CCSDS web site.

One other thing we should consider at the Athens meeting is the Real-Time
Publish-Subscribe (RTPS) protocol standard that is now being advanced within
OMG.  This is the only effort that I know of, besides AMS, to establish an open
standard wire protocol for asynchronous message exchange.

RTPS is already an IEC standard, but unfortunately you have to pay money to read
the specification: $217 at the ANSI Web Store
(http://webstore.ansi.org/ansidocstore/product.asp?sku=IEC%2FPAS+62030+Ed%2E+1%2
E0+en%3A2004) or 292 Swiss Francs at the IEC Web Store
(http://domino.iec.ch/webstore/webstore.nsf/artnum/033452).  As near as I can
tell without buying the book, RTPS is essentially a formal description of the
NDDS commercial messaging product that has been marketed by Real-Time
Innovations, Inc. (http://www.rti.com) for about ten years.

NDDS is designed for hard real-time industrial process control applications, and
I have some doubts that it would satisfy the proposed Requirements for AMS: I
don't think it's likely to be particularly delay- and disruption-tolerant, there
isn't any obvious support for the email-like message staging that Roger talked
about for MOIMS, I didn't see any easy way to partition the message exchange
continuum, it seems pretty firmly tied to UDP/IP without any obvious way to run
it over alternative transmission systems such as message queues or even TCP,
etc.  All the same, we should give it a serious look.  I know that the NASA
Constellation project, at least, is predisposed to going with standards coming
out of the commercial world in preference to standards developed by national
space agencies, so if RTPS can be stretched to address the requirements it will
probably be warmly received.

Scott

_______________________________________________
Sis-ams mailing list
Sis-ams at mailman.ccsds.org
http://mailman.ccsds.org/cgi-bin/mailman/listinfo/sis-ams