Re: old GSE idea

[Sean Doran <smd@ab.use.net>]

On Thursday, Apr 17, 2003, at 18:58 Europe/London, Christian Huitema 
wrote:

> Even if we solved MAC address collision on a single link, you have to
> remember that a host can be multi-homed to several interfaces.
Or virtual/pseudo-interfaces, for that matter.  Hosts can do tunnels, 
VLANs, you name it.

> Whatever
> we do to support the survival of TCP connections should also work when
> the traffic moves from interface A to interface B, e.g. from 802.11 to
> GPRS
The case of moving traffic from a single working interface
to a different single working interface is in many ways easier
than moving traffic across multiple working interfaces simultaneously.

We need to support hosts which are multiply connected within
the same local network infrastructure concurrently (and deal with
failures of some/all of those connections), as well as those which
are simultaneously connected to entirely different networks with 
entirely separate
and possibly mutually hostile administration (and deal with failures of 
some/all
of those connections), and just for fun a mix of the two.

It would be interesting to see if we could be more ambitious and
try to deal with movement/multihoming of things with finer granularity.

The multiported host model is amenable to virtualization, but do we
want to/can we have an identifier bound to a particular *service*
that may migrate around a network independently of other services
which are, at any given moment, sharing a host (virtual or real) with 
it?

Can we/do we want to aggregate several services (or clients thereof)
and have that aggregate be mobile/multihomed?   Yes, because
that's fundamentally what a host is, but can we/do we want several
*independent* aggregates of things within a given host?   Sure: that's
an interesting way of doing virtualization of hosts.   But what about
with granularity at a level other than a host or virtual host?   That 
is,
(assuming process migration is feasible) can I move my
voice over ip service, AIM,  ssh sessions, X server, and so forth
from one part of the network to another without moving the host
these are sharing with an IMAP service, ntpd, and so on?

Most of these questions go to the semantics of the lower-order bits,
how they bind with one or more sets of higher-order bits, and how
they are acquired initially.

	Sean.