On 22-feb-04, at 16:41, Kurt Erik Lindqvist wrote:
We also had this discussed sometime during the autumn and from what I
remember, most seemed to think that mobility and mutli6 required
solutions with different paramters - to the point where the solutions
would be different solutions.
The differences are that mobility needs to be able to change addresses
on short notice, unlike multihoming. With multihoming, there is the
whole selection/failover issue, which is a no-brainer in mobility.
What they could/should share is a way to make transport sessions
survive address changes, and preferably also the security mechanisms.
The problem is that current MIPv6 doesn't do much that is usable for
multihoming, especially not if we rule out the extra "this is the real
address" header because it adds too much overhead.