Cell Relay Archive

Cell Relay Retreat>List Archive>month:1998-Feb> msg00072



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

Re: MPLS paradox

  • From: holdrege@eisner.decus.org (Matt Holdrege)
  • Date: Sat, 21 Feb 1998 05:38:00 GMT

>> On Mon, 09 Feb 1998 13:59:55 +0100, Mike Vogeleer
>> <Mike.Vogeleer@intec.rug.ac.be> wrote:
>> 
>> >MPLS will create a full mesh between all the border routers (each border
>> >must have a VC to every other border router and these VCs cannot be
>> >merged because the switches are not merge capable). I think we all know
>> >that a full mesh means trouble.

I must be missing something. The problems that I know about full meshing are
two. In first generation ATM switches, PVC's are manually created and you can't
create too many of them.

But today's switches can handle up to a million VC's and they can (especially
with MPLS) be created by software.

Since we are talking about protocol managment traffic only, what trouble is
there?