The MPLS-OPS Archive

[Ajay Simha <asimha@cisco.com>]

On Tue Sep 30 21:30:08 2003, sthaug@nethelp.no wrote:
> > Can anyone help me with a small problem.
> > 
> > I have just started on a report  on " Conventional IP routing vs MPLS label
> > switching" with a particular reference to table lookup delay.
> > 
> > This is where I have a small problem which I sincerely hope you somone would
> > take the time to help me out.
> > 
> > Here it is:
> > 
> > I've got a Pentium IV 3GHz machine to be used as a core router maintaining
> > 1000 entries in its routing table, and a visit to an entry in the routing
> > table takes 1 ms.
> > 
> > My main query is:
> > 
> > "How would you calculate the table lookup delay for an incoming packet if
> > conventional IP routing is used, and what would the table lookup delay be if
> > MPLS label switching is used."
> 
> That may be a relevant question for a thesis or a report - but in real
> life, it's a mostly uninteresting question.
> 
> All high speed routers use ASIC-based forwarding or similar these days,
> and these boxes are engineered such that IP address lookup and MPLS
> label lookup run in hardware at about the same speed (the difference is
> small enough to be ignored). Vendors no longer claim that MPLS results
> in faster forwarding.
> 
> If you look at it from another angle - IP is still a huge market, and
> if any router was significantly slower at IP forwarding than at MPLS
> forwarding, it would have a competitive disadvantage. I don't know of
> any boxes that are MPLS-only (but would be interested to hear about
> such).
> 
> > Further as I understand, MPLS label switching might not make routing faster
> > but definitely efficient, in terms of reduced route lookup time. If one has
> > large number of routes in the forwarding table and next hop for all of them
> > is same, I understand that one single label replaces this entire table.
> 
> That doesn't seem to be the case for the vendor I'm most familiar with
> (Cisco):
> 
> #show mpls forwarding-table
> Local  Outgoing    Prefix              Bytes tag  Outgoing   Next Hop
> tag    tag or VC   or Tunnel Id        switched   interface
> 19     Pop tag     10.65.96.16/30    0          PO2/0      point2point
> 20     Pop tag     10.65.192.227/32  0          PO2/0      point2point
> 21     Pop tag     10.65.128.32/29   0          PO2/0      point2point
> 22     12332       10.65.96.4/30     0          PO2/0      point2point
> 23     12333       10.65.192.229/32  0          PO2/0      point2point
> 24     12304       10.65.112.44/30   0          PO2/0      point2point
> 25     12305       10.65.112.32/30   0          PO2/0      point2point
> 26     12306       10.65.112.24/30   0          PO2/0      point2point
> ...
> 
> Here we have lots of prefixes with the same next hop, but different
> labels.

In case of Internet routes, mid-points would only hold labels for exit-points
instead of 150K routes. Same is true for L3VPNs.

-ajay
> 
> 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


"From listening comes wisdom and from speaking repentance."

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