TLA request 'for multihoming' (was: pTLA request SSVL)

[Michel Py]

>>> Pim van Pelt wrote:
>>> From a technical point of view, there is no need for multihoming.
>>> An ISP (look at the thread I started previously) can maintain
>>> redundant uplinks to the Internet, multiple peering routers,
>>> multiple transit carriers, etcetc, without the need for the
>>> customer to have its prefix announced by more than one ISP.
 
>> Michel Py wrote:
>> This does not address:
>> 1. The desire of the customer not to be held hostage by the ISP.
>> 2. The performance requirement that some customers need to have
direct
>> transit from a large number of tier-1.

> Pim van Pelt wrote:
> Both are not technical, but administrative. Especially the first one.

I would call the first one political, but the second one is technical, a
matter of getting directly hooked to the major backbone your customer is
hooked to as well.


> Joel Baker wrote:
> 3. The number of large ISPs now filing Chapter 11 and turning off
>    their networks.
> For mid-size players, #3 is far more crucial than #2, and is really
> just a subset of the causes for #1.

I agree.

> Pim van Pelt wrote: 
> (eg, multihoming itself can be done easily with IPv6 also, but
> aggregation rules forbid it, as could they forbid things in the
> IPv4 world)

You are twisting words. What you are saying is that v6 multihoming could
be possible if we changed the rules and that v4 multihoming could be
impossible if we changed the rules. The situation today is that
multihoming can be done and is being done in v4, and does not exist in
v6. There is no way the v4 rules could change at this point.

Michel.