Re: AW: Multicast in MPLS/BGP IP VPNs

[Chatschik Bisdikian <bisdik at us.ibm.com>]

Dear Nic,

We totally share your view on this point.
The draft has done an excellent job in consolidating multiple proposals
for multicast VPNs under a coherent framework. To accelerate the progress
along the path set by this framework, our proposal wants to place preferential
weight to those portions of the framework that bear as much resemblance
as possible to procedures that have been established and standardized for
unicast 2547 VPNs. In view of this, leveraging BGP (which is an integral
part of multicast VPNs anyway) as much as possible seems quite a reasonable
starting point.

Dr. Chatschik Bisdikian 
IBM T. J. Watson Research Center
19 Skyline Drive, M/S 3S-B34
Hawthorne, NY 10532, USA
tel#: +1 914 784 7439 
fax#: +1 914 784 6205
e-mail: bisdik at us.ibm.com

"Leymann, Nicolai" <Nicolai.Leymann at t-systems.com>

10/04/2005 10:37 AM

To	Chatschik Bisdikian/Watson/IBM at IBMUS, l3vpn at ietf.org
cc	erosen at cisco.com, John.E.Drake2 at boeing.com, rahul at juniper.net
Subject	AW: Multicast in MPLS/BGP IP VPNs

Hi,
 
> a) Be reduced to a minimum number of possible alternatives, and
> b) Clearly spell out required and optional alternatives.
I think thats a major issue with this draft. The draft currently
defines various options and possibilities for the different points
(like encapsulation, routing, discovery, etc.). This also leaves
a wide variety of options for implementations. Therefore I see
the risk with implementations which are not interoperable (even
if those implementations are compliant with the draft). From my
point of view the draft should clearly state which options are 
required for an implementation.

  best regards

     Nic