[88320] in North American Network Operators' Group
Re: MPLS vs PTP
daemon@ATHENA.MIT.EDU (Jon Lewis)
Mon Jan 30 13:43:05 2006
Date: Mon, 30 Jan 2006 13:42:19 -0500 (EST)
From: Jon Lewis <jlewis@lewis.org>
To: Andrew Staples <andrews@ltinet.net>
Cc: nanog@merit.edu
In-Reply-To: <001601c625cb$8298f950$800101df@SapaInc.local>
Errors-To: owner-nanog@merit.edu
On Mon, 30 Jan 2006, Andrew Staples wrote:
> As we roll out a new network, on one of our links it is remarkably cheaper
> to run a T1 ptp vs. MPLS (running 66% data, 33% voice). Based on comments
> received from this list (much thanks, you know who you are) MPLS
> satisfaction seems to be determined by backbone noc competence, not the
> technology itself. So back to price....if I consider layer1 issues to be
> equal in either scenario, and aggregation/meshing/hardware is not a real
> concern, it seems to me that a correctly configured, directly connected pipe
> would work as well as mpls, with the benefit of local control of my routers
> and owning any incompetence.
Also, the PTP T1 has fewer hops (probably lower latency), fewer points of
failure, fewer ways to break, less complexity, etc. I can't think of any
reason you'd want to go with an MPLS(VPN) solution over PTP solution if
startup and MRC were equal.
----------------------------------------------------------------------
Jon Lewis | I route
Senior Network Engineer | therefore you are
Atlantic Net |
_________ http://www.lewis.org/~jlewis/pgp for PGP public key_________
