The MPLS WG Archive

Cell Relay Retreat>MPLS WG Archive>month:2000-Oct> msg00155



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

Draft Minutes from Pittsburgh

  • From: Alex Zinin <azinin@cisco.com>
  • Date: Mon, 16 Oct 2000 10:30:08 -0700
  • CC: Bala Rajagopalan <braja@tellium.com>, curtis@avici.com, George Swallow <swallow@cisco.com>, John Drake <jdrake@calient.net>, "'Debanjan Saha'" <dsaha@tellium.com>, mpls@UU.NET
  • Organization: Cisco Systems


Agree 100%.

We have 3 techniques that are somewhat related, but yet orthogonal:

1. Link bundling
2. Control channel[s] for link bundles (LMP)
3. IGP flooding optimizations

Link bundling allows to summarize the TE information.
LMP allows to minimize the number of adjacencies per bundle
(note that LMP will not have to be implemented on each type
of links), and flooding optimizations minimize overhead traffic
over parallel adjacencies.

Alex.

Monday, October 16, 2000, 9:05 AM, Curtis Villamizar <curtis@workhorse.fictitious.org> wrote:

> Bala,

> I remember there being consensus at Pittsburgh on making the Kompella
> bundles draft a WG item.

> Use of the flooding optitimization is one means of improving the
> efficiency of the IGP and is a separate issue.

> Both the bundles and the flooding optitimization can stand alone.
> Neither one depends on the other.  If used together they may
> complement each other but use of either one separately is possible and
> useful.

> I don't see that any changes are required to the bundles draft to
> accommodate the flooding optitimization.  If changes are required,
> please point them out.

> Curtis