Re: [ssm] msnip status

[hoerdt Mickael <hoerdt at clarinet.u-strasbg.fr>]

Hi,

MSNIP (or any logical equivalent, let call it protocol X) allows 
ressources saving at the following points of the network :

1. the DR(s).
2. the link connecting the DR to the multicast source host(s).
3. the host(s).

A possible scenario showing that protocol X is (relatively) useful would 
be the following :

. Exhaust the ressources available at any of these three points, 
sourcing multicast
 content without  protocol X.
. Activate protocol X, and replay the same scenario showing that now the 
ressources
 aren't exhausted.

Does it make sense to define a protocol X wich do not have to be 
deployed/implemented
in the routers, using possibily unicasting and/or pooling from the 
source ? As a multicast
application developer/coder, i would say that it depends on the SSM 
multicast service
availability....

Hoerdt Mickaël





Pekka Savola wrote:

> On Wed, 19 Oct 2005, Toerless Eckert wrote:
>
> > I think MSNIP tried to solve a key functionality for the SSM 
> > architecture,
>
>
> Could you elaborate a bit on what you think is this "key 
> functionality" and why exactly it's "key" (e.g., specific scenarios 
> you have in mind)?
>
> I'm having difficulty understanding the (relative) importance of a 
> sender being able to stay quiet if there are no receivers.  The waste 
> of bandwidth and processing power at the DR (who'd just discard the 
> traffic) doesn't seem to be sufficiently concinving at least in the 
> scenarios I have in mind.


_______________________________________________
ssm mailing list
ssm at ietf.org
https://www1.ietf.org/mailman/listinfo/ssm