All things related to multicast

[Greg Shepherd <>]

On Tue, 28 May 2002, Marshall Eubanks wrote:

> As we bring up video channels, we plan to mirror them in SSM.
> If the various _players_ as well as the OS's will support them, we
> will drop the ASM support. My guess is that this will take a year to
> happen.

Marshall,

I really think you're doing the right thing here. You may also want to
consider making the SSM stream a higher quality feed as an incentive. In
fact, this is similar to some ideas to incent mcast deployment I've tossed
around with others; create a short 'demo' unicast stream at a much lower
bitrate, a link to a screen-shot of the higher bitrate mcast feed, and end
the demo with some hype-text to create a buzz.

"I want my McastTV!!"

Sure, and we'll get David Bowie to give the plug. ;-)

Greg

> Regards
> Marshall
>
> John Zwiebel wrote:
>
> > Liming Wei wrote:
> >
> >>But lets lighten up a bit and remember that there is
> >>IGMPv3/SSM/PIM-SM, making all these issues moot. In the quest for a
> >>better or more easily implementable protocol, we should not (do not
> >>need to) break things that were working before.
> >>
> >>-Liming
> >>
> >
> >
> > Liming:
> >
> > One would think this would be an easy sell.  How many OS do you
> > know that have IGMPv3 installed as standard?  Windows XP is the
> > answer.
> >
> > Yes, you can build a FreeBSD kernel with it, and supposedly there's
> > a Linux patch, but they aren't there.  In fact, the Linux patch
> > is (according to an email exchange I've had on Linux.org)...
> >
> >    Last time i saw IGMPv3 patches was some from sprint labs
> >    posting and if i recall they were not SMP safe.
> >
> > Then there has been the discussion I've had with some I2 folks
> > who like their application just the way it is and don't want
> > to migrate to SSM.
> >
> > And... there is the question of what to do with an IGMPv3 (S,G)
> > report received for a non-SSM group (ie outside of 232/8).  Either
> > I've chosen some incorrect terms resulting in a huge misunderstanding,
> > or I'm not getting back responses that make sense to me.  Some
> > individuals are adamant that such a join should be interpreted by
> > PIM as a (*,G) join or should be ignored.
> >
> > Let me state again, that there is no reason a last-hop router should
> > be prevented from establishing a shortest-path tree no matter what
> > group range is being used.  There is no reason to not include
> > the source address in the SAP description of the session so that
> > an "SSM-aware" application can join directly to the source by-passing
> > the shared-path.  (There may, of course, be administrative reasons,
> > but those should be interpreted by the network manager in control
> > of his network.)
> >
> > NOTE: I said "SSM-aware", meaning an application that can
> > take advantage of IGMPv3.
> >
> > Anyway, the discussion is moot unless we "finish" the job and
> > get IGMPv3 into Linux and FreeBSD (from which I understand it
> > will 'automagically' appear in Darwin -- ie MacOSX).  It appears
> > that app developers think it is "just too hard".  (I don't mean
> > that in a nasty way, rather that the impression I have
> > developed is that app developers don't want to deal with
> > something that they are not sure is useful.  Good point.)
> >
> > So, if SSM is going to go anywhere, we need to do a much
> > better job of selling it.
> >
> > z
> > _______________________________________________
> > pim mailing list
> > 
> > http://catarina.usc.edu/mailman/listinfo/pim
> >
>
>
> --
>
> T.M. Eubanks
> Multicast Technologies, Inc
> 10301 Democracy Lane, Suite 410
> Fairfax, Virginia 22030
> Phone : 703-293-9624       Fax     : 703-293-9609
> e-mail : 
> 
> http://www.multicasttech.com
>
> Test your network for multicast :
> http://www.multicasttech.com/mt/
>   Status of Multicast on the Web  :
>   http://www.multicasttech.com/status/index.html
>
> _______________________________________________
> pim mailing list
> 
> http://catarina.usc.edu/mailman/listinfo/pim
>