The MPLS-OPS Archive

Cell Relay Retreat>MPLS-OPS Archive>month:2005-Apr> msg00059



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

RE: MPLS vs IP forwarding speed..

  • From: "Andrew Walding" <andyw@cellstream.com>
  • Date: Thu, 21 Apr 2005 11:43:10 -0500
  • Resent-Date: Thu, 21 Apr 2005 13:16:45 -0400
  • X-AntiAbuse: This header was added to track abuse, please include it with any abuse report
  • X-AntiAbuse: Primary Hostname - mickey.1host.com
  • X-AntiAbuse: Original Domain - mplsrc.com
  • X-AntiAbuse: Originator/Caller UID/GID - [0 0] / [47 12]
  • X-AntiAbuse: Sender Address Domain - cellstream.com
  • X-PopBeforeSMTPSenders: andyw@cellstream.com,andyw@totalcorporatecompliance.com,awalding@totalcorporatecompliance.com
  • X-Scanned-By: MIMEDefang 2.45

I'm sorry - I can't help myself...and have to step in here...
The IETF RFC's contain all the actual reasons that what is now called MPLS
generically evolved from cell switching or IP switching or Tag Switching -
in other words the challenge of performance of IP forwarding over ATM
fabric.  These papers identify the reasons this technology developed.  Much
of the rest of the reasons enumerated here and in other places are great
lore and marketing spins on the actual technology developments.

Most of the pertinent RFC data and papers I refer to can be found around the
1900-2100 RFC range.  There are many solutions and problem statements in the
Ipsilon, NEC, IBM, Cisco, and Toshiba papers.  This all began in the
1994-1995 time frame.

Have fun with the actual history.

______________________________
Andrew Walding, President
CellStream, Inc.
9812 Zembriski Drive, Plano TX  75025
Tel: 972-747-0007   Fax: 972-747-1130
www.cellstream.com
 

> -----Original Message-----
> From: sthaug@nethelp.no [mailto:sthaug@nethelp.no] 
> Sent: Thursday, April 21, 2005 11:11 AM
> To: mpls-ops@mplsrc.com
> Subject: Re: [MPLS-OPS]: MPLS vs IP forwarding speed..
> 
> > > http://cell.onecall.net/mhonarc/mpls/1997-Jul/msg00059.html
> > > is another one which says it is faster
> > 
> > Well, 1997 is the "stone ages" for MPLS, and lots of strange claims
> > were made.
> 
> Followup: In the message from Grenville Armitage referred to above:
> 
> - #1 refers to forwarding based on caching the forwarding information.
> It is not particularly relevant to today's *ASIC* based forwarding
> engines. I believe it's not particularly relevant to CPU based
> forwarding either: You would not *use* such a router precisely because
> performance would be abysmal.
> 
> - #2 refers to forwarding based on an ATM cell fabric. This, too, is
> not particularly relevant today.
> 
> So, my basic claim is that the world has moved on, and claims made in
> 1997 about an MPLS forwarding speed advantage simply aren't relevant
> today.
> 
> Steinar Haug, Nethelp consulting, sthaug@nethelp.no
> 
> -------
> The MPLS-OPS Mailing List
> Subscribe/Unsubscribe:  http://www.mplsrc.com/mplsops.shtml
> Archive: http://www.mplsrc.com/mpls-ops_archive.shtml
> 
> MPLScon 2005 - May 16-19, NYC, NY
> http://www.mplscon.com/
> 
> 

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

MPLScon 2005 - May 16-19, NYC, NY
http://www.mplscon.com/