The MPLS-OPS Archive

Cell Relay Retreat>MPLS-OPS Archive>month:2002-Jan> msg00248



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

Re: MPLS/"Best Effort" Bandwidth Contention

  • From: Santiago Alvarez <saalvare@cisco.com>
  • Date: Wed, 30 Jan 2002 09:04:31 -0800
  • CC: "'mpls-ops@mplsrc.com'" <mpls-ops@mplsrc.com>
  • Resent-Date: Wed, 30 Jan 2002 13:30:29 -0500
  • To: stephen mullaney <stephen.mullaney@parc-technologies.com>

BW guarantees in the forwarding plane are implemented with a queuing
infrastructure and a scheduler that allocates BW to those queues.  If
you differentiate the traffic correctly at every node and map it to the
correct queue, the scheduler will guarantee a minimum BW allocation to
the queue regardless of the offered load to other queues.

SA
--

> stephen mullaney wrote:
> 
> Hi there,
> 
> This may be a Cisco specific question, but....
> 
> For a given link which carries both a MPLS guaranteed tunnel and some
> "best effort" non-MPLS traffic how is the bandwidth constraint for the
> tunnel enforced.
> 
> eg if the tunnel is 20M and the link is 100M, I assume that if no
> other traffic is on the link, the tunnel can use more than the 20M
> assigned to it.  But when "best effort" traffic starts to traverse the
> link, how are the two types of traffic treated. Is the MPLS traffic
> pushed-down to its bandwidth guarantee as the best-effort increases?
> Do both types of traffic get the same treatment/queuing? What is the
> algorithm for differentiation? etc?
> 
> Thanks
> Stephen

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