Re: [v6ops] [GROW] Deaggregation by large organizations

[Ray Hunter <[e-mail]>]

Russ White wrote:
>> Injecting an aggregate as a point of last resort:
>>
>> I think this can be done today and probably is done today. But a document
>> describing how to do it would probably be helpful. I'm thinking along the
>> following lines:
>>
>> The AoLR (Aggregate of Last Resort) service would entail a service
> provider
>> announcing the aggregate without necessarily providing connectivity
> towards
>> all the places announcing more specifics covered by the aggregate. So if
> ISP A
>> announces the AoLR and ISP B provides connectivity to a more specific, ISP
> C
>> would send traffic to A as per the aggregate and then A would immediately
>> hand it over to B.
>
> Or perhaps something simpler would work here -- bounding longest match.
>
> :-)
>
> Russ
>

No.
Now we've "solved" the address space issue, we focus back on routing 
table size angst. History repeats.

My (tiny) datapoint on multinationals is that they're simply grabbing 
/32's (or as big as possible space) from multiple registries as soon as 
possible: remembering how much of a battle it was in IPv4 to get PI 
space properly routed, and assuming that this same discussion would 
resurface in IPv6, and that maximum prefix length would once again 
become a hot topic..... short prefixes are "better". Get 'em now while 
you can.

This is natural behaviour, unless there's a less blunt instrument 
applied to managing BGP routing table size than simply restricting 
maximum prefix length per peer. e.g. RIR charge being dependent on the 
number of global routing slot entries, rather than purely on address 
space allocation size.

And yes, of course, de-aggregation is a potentially serious DOS vector.

Meanwhile, renumbering is still hard.

-- 
Regards,
RayH