[AG-TECH] Fwd: New Draft - RTCP for SSM

[Robert Olson]

fyi.

>Sender: julian at research.att.com
>Date: Tue, 06 Mar 2001 10:58:48 -0800
>From: Julian Chesterfield <julian at research.att.com>
>Organization: AT&T Labs
>X-Mailer: Mozilla 4.76 [en] (X11; U; FreeBSD 4.2-RELEASE i386)
>X-Accept-Language: en
>To: rem-conf at es.net
>Subject: New Draft - RTCP for SSM
>X-Mailing-List: <rem-conf at es.net>
>X-Loop: rem-conf at es.net
>
>Please find a new draft available online at:-
>
>http://search.ietf.org/internet-drafts/draft-chesterfield-avt-rtcpssm-00.txt
>
>This is a proposal for the adaption of RTCP to operate in a Source
>Specific Multicast Session where there is no 'many to many'
>communication for which the protocol was originally designed. This is my
>first shot at writing drafts, so I suspect there are a number
>innaccuracies! However, the main points which are being proposed are:-
>
>- Receivers unicast reports to the source/sender in the session.
>- The source is responsible for distributing the reports on the SSM RTCP
>channel to ensure the scaling algorithm still works.
>- The source has 2 options for distributing the data. The first is a
>Sender Report in which the Report Blocks correspond to the individual
>reports from the receivers. Since the source is the only member
>generating SRs, I think this should be fairly clear to the receivers.
>The second is a proposed new packet format named Sender Summary Report
>(SSR) in which the redundant Timestamp information is removed to enable
>less packet size overhead.
>
>I would appreciate comments and feedback on the draft. My personal
>reason for pushing ahead with this is that I am working on Multicast
>Network Monitoring tools, and RTCP data is a key part in providing
>information. I want to be able to retrieve that information in Source
>Specific Multicast sessions as well.
>
>Many Thanks,
>
>Julian Chesterfield
>AT&T Labs Research,
>Menlo Park,
>CA