Re: On BGP and VPLS

Alex,

[clipped...]

>  3. Aggregation of information for large-scale operation
> 
>    When distributing information among a large number of systems, it
>    is important to be able to aggregate information as it travels
>    further ahead to ensure scalability of the system. In routing this
>    is achieved by summarizing a set of prefixes and announcing them as
>    a less specific prefix. For example, AS'es in the Internet do not
>    exchange granular IP prefixes visible inside IGPs, but instead send
>    each other aggregate prefixes via BGP.
> 
>    It is not clear to me how, given the format of the NLRI field, VPLS
>    information can be aggregated using the proposal in the document.

It should be abundantly obvious to an informed reader that information
aggregation/abstraction is just *one* way to support large-scale
operation. But it is certainly *not the only way* possible.

Section 15 of draft-ietf-ppvpn-rfc2547bis-03.txt and Section 13 of
draft-ietf-ppvpn-as2547-01.txt describe how BGP/MPLS VPNs supports
large-scale operation in the absence of aggregation/abstration of
VPN routing information.

It should be abundantly obvious to an informed reader that use of
BGP for autodiscovery (draft-ietf-ppvpn-bgpvpn-auto-03.txt), as
well as for VPLS autodiscovery and label distribution utilizes
exactly the same mechanisms as BGP/MPLS VPNs to support large-scale
operation.

>  The above gives me a very uncomfortable feeling that the proposal
>  is stretching BGP to perform functions it was not designed for.

IP wasn't designed to support CIDR.
TCP wasn't designed to carrying routing information (BGP).
RSVP wasn't designed to establish explicitly routed LSPs.
etc...

And, bwt how do *you* know what BGP was designed for ? If one really
wants to know what BGP was (and was not) designed for, then perhaps
one should ask the folks who designed BGP.

Yakov.