The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2002-May> msg00252



[Date Prev][Date Next][Thread Prev][Thread Next]  
  [Date Index][Thread Index][Author Index][Subject Index]

Basic LDP Question

  • From: Vijay Bollapragada <vbollapr@cisco.com>
  • Date: Fri, 31 May 2002 14:05:36 -0400 (Eastern Daylight Time)
  • cc: "'mpls@uu.net'" <mpls@UU.NET>, "'ppvpn@ppvpn.francetelecom.com'" <ppvpn@ppvpn.francetelecom.com>
  • X-X-Sender: vbollapr@mayo.cisco.com

In the 2547 case, typically u have 2 labels, the vpn_label (bottom label)
and the top label is the label to get the packet to the far end PE, Now,
LDP is used for this top label.

I think u are questioning why use LDP for this bottom_label and why not
simply encapsulate the MPLS packet in IP to get the packet to the next_hop
PE. If this is your Q, You are absolutely correct and this is what the
following draft talks about

http://search.ietf.org/internet-drafts/draft-ietf-ppvpn-gre-ip-2547-01.txt

IMHO, this is certainly do-able but MPLS in the core give us the ability
to do things like Traffic Enginnering which are quite a challenge over
pure IP.

Thanks,
Vijay



On Fri, 31 May 2002, Shahram Davari wrote:

> Hi,
> 
> What problem a mp2p LDP-based MPLS network is trying to solve (like the one used in RFC2547) that can't be done with pure IP forwarding?
> 
> Note: Don't tell me fast forwarding, because we can do fast IP lookup as well.
> 
> Yours,
> Shahram
> 
>