Robert, > In the interest of saving time tomorrow I would like to ask for some > clarification reg how large corporate enterprise sites (those who run full > table of BGP today) would auto discover ITRs ? > > Their network maybe multi-as as well as multi-area. The answer from the > audience of LISP tutorial was that EIDs-subnets would be carried in those > networks in BGP to attract traffic to ITRs. But on the other hand we know > that there can be much more EID subnets (I think I heard 3 fold estimate) > then there is today IP prefixes in full BGP table ... not too mention 10^10 > long term number.
As you point out, today an enterprise finds its "best
exit" by using the (BGP) routes that it learns from its
various providers and distributing internally them with
iBGP. If the ITRs are sitting on the edge of the site's
network, this won't work since packets will be routed
internally following default. A packet must follow
default because there no information about the best exit
for a given EID prefix.
So in order to preserve the behavior you are describing,
the packet needs to be routed over the RLOC space as
early as possible. This corresponds to placing the ITRs
deeper in the site (closer to the source hosts). Since
the ITR will encapsulate the packet (and the outer header
destination address will be a locator), it will find the
best exit just as it does today.
> What would be the benefit and driver for such large corporate sites to
> install ITRs ?
Finer-grained control over exit policy without having to
propagate global state inside your enterprise.
Dave
signature.asc
Description: Digital signature
