Re: draft-ietf-multi6-multihoming-requirements-06.txt

[Iljitsch van Beijnum <iljitsch@muada.com>]

On woensdag, jun 11, 2003, at 21:53 Europe/Amsterdam, Randy Bush wrote:

> Let's see: load balancing is not a realistic requirement.  For the
> architecture to be automatic, we have to be dynamic and we've
> proven that we don't know how to do that yet.  Even to give people
> some tools seems to lead to abuse.  Who controls the balancing?
> The source?  The destination?  Don't say 'both'.  No packet can
> serve two masters.
No load balancing whatsoever means having a primary connection and a 
backup. For most multihomers paying for bandwidth you only get to use 
less than 1% of the time is a non-starter.

> Performance.  Exact same arguments since the only way to affect
> performance is to load balance.
Having a fast link, adding a slow one and having performance limited by 
the slow one is again unaccaptable.

And it's not like providing some basic load balancing is a huge deal. A 
simple preference value for each of multiple AAAA records or some such 
would go a long way.

> The real requirements are:
> 	- Provide multihoming
> 	- Provide scalable routing
> 	- Transport survivability
The real real requirement is: it must be good enough that people will 
want to trade in their IPv4 provider independent address space for 
this. If it isn't, either they'll stay in IPv4 or they'll take the IPv4 
multihoming mess with them during the transition to IPv6.