The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-Mar> msg00362



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

tspec table

  • From: Juha Heinanen <jh@lohi.eng.telia.fi>
  • Date: Tue, 28 Mar 2000 08:41:54 +0300 (EEST)
  • Cc: mpls@UU.NET, Arun Viswanathan <arun@force10networks.com>, cheenu@tachion.com

Thomas D. Nadeau writes:

 >          Please be more specific as to how the cr-ldp
 > parameters differ from those currently in the MIB.
 > Please give specific examples.

cr-ldp has the following traffic parameters:

- frequency
- peak data rate (same as mplsTSpecMaxRate)
- peak burst size
- committed data rate (same as mplsTSpecMeanRate)
- committed burst size (same as mplsTSpecMaxBurstSize)
- excess burst size

so it appears that cr-ldp traffic parameters are a superset of rsvp
traffic parameters and we might be able to solve the problem by renaming
the current table and parameters using generci names and by adding the
missing three parameters.

 >          The LSR-MIB models how labels are swapped (among
 > other things), which is why I believe that name was
 > chosen. With the MIB currently in WG Last Call, I do not
 > believe that we will be changing its name at the 11th
 > hour.

according to the terminology section of the architecture i-d, mpls node
is a node that is dealing with labeled packets and can POSSIBLY also
forward l3 packets, whereas lrs is an mpls node that also does forward
l3 packets.  that is why i would consider the mpls node mib a more
appropriate name for the i-d.

-- juha




  • Follow-Ups: