The MPLS-OPS Archive

Cell Relay Retreat>MPLS-OPS Archive>month:2001-Mar> msg00127



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

Re: Questions about MPLS

  • From: Eric Osborne <eosborne@cisco.com>
  • Date: Sun, 18 Mar 2001 20:48:44 -0500
  • Resent-Date: Sun, 18 Mar 2001 21:59:45 -0500
  • To: "Tim A. Irwin" <tirwin@bellsouth.net>, mpls-ops@mplsrc.com
  • User-Agent: Mutt/1.2.5i
  • X-GPG-Fingerprint: 6412 0836 E440 B3EA 980C 4951 611E 1819 2E71 8562

Folks-

  The point I've been trying to make is that there's more than one way
to skin a cat.  IPSec is useful for some things, MPLS-VPN is useful
for others, and there's functionally a significant degree of overlap
between the two.  I have no problem with people saying "I don't like
MPLS-VPN", but I'm tired of "MPLS VPN sucks because I say it sucks".
That's my only objection, really.  

  When talking about service provider architectures, there are two
rules.

1) they all look pretty much the same
	- 2 or 3 layers of routers in a POP, POPs connected to other
POPs, connections to customers and external peers.  BGP, and generally 
speaking, no default route.
 
2) they're all different
	- POS vs. ATM WAN links, OSPF vs. IS-IS vs. (a little) EIGRP,
MLPS vs. IP, MPLS-VPN vs. IPSec, MPLS-TE vs. ATM, POS vs. DPT/SRP
vs. GigE in the POP, Coke vs. Pepsi, QoS or no QoS, oversubscribe
ratios, number of functional types of routers in a POP, access speed,
SLAs, etc...


For every one of the points in #2, I bet there's at least one person
on this list that's rabidly on one side, and one that's rabidly on the
other.  All I'm trying to point out is that the existence of MPLS-VPN
is just another option of something you can run.  Don't like it?
Don't run it.  But the fact remains, some people are using it and like
it.  And there's some very smart people who seem to think it's useful;
at least one of them works for a company that it's not fashionable to
hate. :)

As far as the 'barrier to entry' bit, that seems like a reach to me.
Sure, having to do MPLS-VPN if you want to pick up a certain set of
customers is more work than not doing MPLS-VPN.  But so's doing a POS
interface.  GR-253 is almost 800 pages long; RFCs 3036, 2547, and
2328, and 1771 together are little more than half that length.  So if
MPLS VPN is a barrier to entry, so are SONET, IP, SNMP, and all that
other stuff I hear is all the rage nowadays.

Can't we all just get along? 




eric

-------
The MPLS-OPS Mailing List
Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
Archive: http://www.mplsrc.com/mpls-ops_archive.shtml