Re: [RRG] Re: Fast and sparse mapping?

[Brian E Carpenter <brian.e.carpenter@gmail.com>]

On 2008-09-21 16:52, David Conrad wrote:
> Brian,
> 
> On Sep 20, 2008, at 7:08 PM, Brian E Carpenter wrote:
>> I don't think we need to design for a world where most domestic
>> subscribers
>> are multihomed, or care in the least if they get a new IP address
>> each time they connect.
> 
> I would classify this as a failure of imagination. If RRG is indeed
> aiming at 10 to 20 years in the future, I would be quite astonished if
> an underlying assumption is that the way the Internet is today is the
> way it will be one to two decades in the future, just bigger.

However, many people here seem to be assuming that current operational
practice will be just the same, specifically the attachment to
unscalable techniques such as PI addressing and boundary firewalls.
If we take those as givens, we're forced into a conservative mode
of thinking.

> My assumption is that as people become more and more dependent on
> Internet connectivity for their day-to-day lives, the less interested
> they will be in periodic outages.  When the myriad devices in your house
> depend on being connected to the Internet and you have a wide variety of
> layer 2 technologies over which to connect (e.g., WiMax, FTTH, Cable,
> DSL, and their successors), I fully expect most domestic subscribers to
> be multihomed.  Actually, I expect it to be worse than that -- I can
> very easily imagine multi-homed PANs connected to multiple providers via
> cell phone-as-router-equivalents.

Agreed.

> 
> As such, in my opinion, a design that does not anticipate (or at least
> cannot scale with) massive multihoming would be a waste of time.

We've had that design for almost 15 years - it's called PA, with
multiple prefixes per PAN. As I already said, I don't think the
mass-market users even know what an IP address is, and certainly
don't have any attachment to PI. PI and its consequent problems
are for larger networks.

I think this is very do-able: PI for ten million larger sites,
with LISP-type mapping if needed, and multi-prefix PA for
maybe ten billion consumer nodes, PANs and MANETs.

     Brian

--
to unsubscribe send a message to rrg-request@psg.com with the
word 'unsubscribe' in a single line as the message text body.
archive: <http://psg.com/lists/rrg/> & ftp://psg.com/pub/lists/rrg