The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-May> msg00396



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

MPLS Performance analysis.....

  • From: Jeremy Lawrence <jlawrenc@cisco.com>
  • Date: Thu, 01 Jun 2000 11:58:59 +1000
  • Cc: "'Sean Doran'" <smd@ebone.net>, "'Martin Cooper'" <mjc@cooper.org.uk>, <mpls@UU.NET>

At 16:09 05/31/2000 -0400, Brijesh Kumar wrote:
[...]

>When MPLS work started the enhancing the speed of IP route look up was a key
>motivator behind the technology.  

Not really, except in the mind of certain marketers who yelled
"[Company X] is only doing MPLS because they can't get their
routers working fast enough". Company X, in the meantime, knew
that MPLS was only a marginal performance improvement over a
good IPv4 longest-prefix match implementation. :)

There is one grain of truth in that fast IPv6 longest-prefix match
algorithms are harder to come by, and MPLS was considered as
a solution to that problem.

Jeremy