Re: Computing Traffic Matrix from netflow data

[Jennifer Rexford <jrex@research.att.com>]

Jing,

> Is there any work on deriving traffic matrix for OSPF area 0?  

If each area is a region, then you could aggregate Netflow meaurements
for traffic entering the region from the edge links and aggregate the
BGP next-hop IP addresses based on the OSPF area of the outgoing router.
That is, you could use the standard approach for estimating the router-
to-router traffic matrix but just aggregate the data for all ingress
routers in the same area and all egress routers in the same area.

> Any work on OSPF optimzation for traffic engineering?

For surveys on this topic, see

  Bernard Fortz, Jennifer Rexford, and Mikkel Thorup, "Traffic
  engineering with traditional IP routing protocols," IEEE 
  Communication Magazine, October 2002.
    http://www.research.att.com/~jrex/papers/ieeecomm02.pdf

  Jennifer Rexford, "Route optimization in IP networks," to appear as
  a chapter in Handbook of Optimization in Telecommunications, Kluwer
  Academic Publishers, 2005.
    http://www.research.att.com/~jrex/papers/opthand04.pdf

and references therein.  The second paper, in particular, has quite
a lot of recent references on optimizing OSPF/IS-IS weights.

-- Jen