On Wed, Oct 27, 2010 at 11:44 AM,  <[email protected]> wrote:
> In einer eMail vom 26.10.2010 20:40:46 Westeuropäische Sommerzeit schreibt
> [email protected]:
>
> I was also a little bit shocked - since I have proposed hIPv4  -when I
> noticed that hierarchical routing is a bad choice...
>
> You should see TARA's advantages wrt hipv4. As routing occurs based on the
> dest. TARA-locator, the dest address may be what so ever: IPv4, IPv6, HIP
>
>
Hi Heiner,

I see the advantages but I think the routing protocol is not the issue
here - IMHO, it is the old fashioned transport protocol lacking a
session layer mechanisms that is causing the pollution at the DFZ. So
one approach is to
a) add a session layer mechanism (such as HIP, ILNP, NBS) to the
current transport protocols
b) the routing architecture is changed slightly so that it better
serves the needs of  multipath transport protocols, used with or
without HIP, ILNP, NBS mechanisms
c) LISP is also creating a sort of session layer, but is handled by
the network devices instead of the endpoints - it is a good starting
point (and needed) since it will speed up the migration phase - but
sooner or later the endpoints' stack  will be upgraded.

Note that this approach doesn't exclude an upgrade of the routing
protocol either - an upgrade can be carried out if there is incentives
for it.

IMHO, the root cause for our problems is the lack of a session layer
in the TCP/IP stack

Patrick
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to