The MPLS WG Archive

[Curtis Villamizar <curtis@fictitious.org>]

In message <001901c2d869$8863c200$3a785b87@mt.att.com>, "Jim Hand" writes:
> Also, neither draft requires end-to-end MPLS between VoIP endpoints.  One of
> the drafts
> (http://ietf.org/internet-drafts/draft-ash-e2e-vompls-hdr-compress-00.txt)
> requires MPLS between the compressor and decompressor.  The other
> (http://ietf.org/internet-drafts/draft-ash-e2e-crtp-hdr-compress-00.txt )oes
> not require MPLS all the way between compressor and decompressor.  Neither
> draft requires that the compressor and decompressor also be VoIP endpoints.
> 
> Thanks,
> Jim Hand


By end to end I meant compression device to compression device.  I did
say gateway to gateway which might imply VOIP device.

The argument I was making was that if the compression device is very
far toward the edge of the network, then the number of LSPs explodes.
If the compression device is moved closer to the core of the network
such that a full mesh of RSVP/TE LSPs are feasible, then opportunity
for multiplexing is enormous.

If you are going to burden the CPU at the "compression device" why not
have it multiplex rather than header compress and get huge improvement
in encapulation efficiency rather than relatively small improvement in
encapsulation efficiency.  This also makes it feasible to more the
compression closer to the edge and maybe into the VOIP gear rather
than a router concentrating VOIP gear traffic.

Thanks again,

Curtis


> > -----Original Message-----
> > From: Curtis Villamizar [mailto:curtis@fictitious.org]
> > Sent: Wednesday, February 19, 2003 11:25 AM
> > To: Ash, Gerald R (Jerry), ALABS
> > Cc: curtis@fictitious.org; raymond zhang; MPLS@UU.net; George Swallow;
> > Loa Andersson; GOODE, B (Bur), ALABS; Hand, James C, ALABS;
> > Dave Cooper
> > Subject: Re: E2E VoIP over MPLS ('VoMPLS') Header Compression
> 
> >
> > Please correct me if I'm wrong but your drafts seem to require end2end
> > MPLS where RTP/UDP/IP/MPLS normally requires just end2end IP.  Since
> > you are providing new RSVP/TE objects the assumption seems to be
> > RSVP/TE MPLS edge to edge.  If traffic engineering is done within
> > regions as can be done with current RTP/UDP/IP/MPLS regardless of
> > whether LDP is also used, that works fine and scales extremely well.
> > E2E RSVP/TE MPLS all the way to each peice of VOIP gear is unlikely to
> > scale well in a single provider and is almost certain to become a
> > severe problem crossing SP boundaries.
> >
> > Curtis