The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2002-Jul> msg00392



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

AW: AW: initiating a discussion on mpls singaling protocols

  • From: David Charlap <david.charlap@marconi.com>
  • Date: Fri, 26 Jul 2002 13:25:33 -0400
  • User-Agent: Mozilla/5.0 (Windows; U; WinNT4.0; en-US; rv:1.1b) Gecko/20020719

Hummel Heinrich wrote:
> As I see it:
> 
> 1) p2p     LSP: in DoD. Is done by RSVP-TE and CR-LDP. With and without Explicit Routing.  
>        
> 2) mp2p    LSP: Is done by  LDP using DU. But without  QoS-/Policy-/Bandwidth- sensitiveness;
>                 Could be done  by explicitly-routed /hop-by-hop routed merging LSPs, starting from the egress.
>                 Is not provided by CR-LDP. Is it provided by RSVP-TE (I doubt)?

RSVP-TE does not support this.  Being a TE protocol, each LSP originates 
from exactly one ingress node.

It should be theoretically possible for multile LSPs to merge (share an 
outgoing label) if their QoS and explicit routes are identical for the 
remainder of the path to the egress, but this is more of an optimization 
than a feature that can be explicitly requested.  I would hesitate to 
call the result of this a mp2p LSP.

> 3) p2mp    LSP (for Multicast): from source towards receivers, upon leaf initiated join requests. Not even started.

RSVP-TE, as specified right now, can handle this.  RFC 3209 does not 
specify a multicast version of the ERO (explicit route object), but I 
think the rest of the RFC can work fine for multicast, if the tunnel 
endpoint address is a multicast group address.

A multicast ERO object has been defined in Saisho Yasukawa's draft, if 
you consider that a requirement for p2mp LSPs.

I don't think there any implementations at this time, however.  I don't 
know which (if any) vendors are currently working on this.

> 4) mp2mp   LSP (e.g for VPLS): Is not provided at all. Note: If n=2, it means the bi-directional LSP. Is not provided at all.
>                                Could be done by explictly routed /hop-by-hop routed LSPs starting from whichever
>                                edge node towards all the other edge nodes. 

I don't know of any work in this area.  I think (but don't know enough 
to be sure) that LDP could theoretically work here, by using multicast 
group addresses as FECs.  But I don't know the protocol well enough to 
know if there are other issues that might make it impossible.

RSVP-TE, as currently defined, can't do this.  Same reasons as mp2p (see 
above.)

-- David