The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-Oct> msg00524



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

VPN solution - White flag ?

  • From: Randy Bush <randy@psg.com>
  • Date: Thu, 26 Oct 2000 10:09:19 -0700
  • Cc: Jim Guichard <jguichar@cisco.com>, Paul Doolan <pdoolan@ennovatenetworks.com>, Barry Hass <BHass@nexabit.com>, erosen@cisco.com, yakov@cisco.com, rnewcomb@ennovatenetowrks.com, mpls@UU.NET, diego@estos.upc.es

> Then perhaps this is not the solution for you.  However, the fact
> that it is not applicable to your particular application does not
> make it worthless (this is to be differentiated from drafts
> documenting architectures that are applicable to no applications).
> 
> Is it inappropriate, in general, for SPs who need to maintain full
> Internet tables in their PEs?  Quite likely.
> 
> Is it inappropriate, in general, for SPs who have nothing but VPN
> customers, each of whom has very few prefixes?  Probably not.
> 
> I don't believe anyone has challenged either MOs or your ability to
> engineer ISP networks, but perhaps viewing everything through your
> ISP goggles is not the most effective way to judge the utility of
> all drafts.

so we agree it is not for isps.  cool.

so, if an enterprise wanting to deploy it is not an isp, then they'll
need their own layer 1 or2 connectivity between all endpoints.  so then
what is the advantage to them of this clever (a pejorative) approach to
old-style layer 2 vpns?

randy