Hello DY,

I agree with your statement (msg07615):

> I don't think decoupling mobility from routing should be listed as a
> design requirement. I don't think we can eliminate the possibility of
> routing inherently providing mobility.

The term "routing" is not defined clearly in:

  http://tools.ietf.org/html/draft-irtf-rrg-design-goals-04

If it means "every active device between the hosts" then this would
mean all the existing routers and whatever new devices, or additional
router functionality, is provided as part of a Core-Edge Separation
architecture (LISP, Ivip or IRON).   Since these additional devices
and functions could be designed to handle mobility in a scalable
fashion, it seems a bad idea to mandate in the design goals that this
not be allowed.

No-one is suggesting that existing routers do a lot more work to
accommodate mobility.  If the goal was simply to protect existing
routers from such an extra burden, it might make more sense, if it
could be shown that this could never be done scalably.

Since the whole project is about scalability, and since some solutions
(in my view, the only ones which could work) involve devices other
than the hosts doing the extra work, all that is required is that the
new and existing devices do their work scalably.

  - Robin


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

Reply via email to