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

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

Resent, since the list's anti-trivia protection interpreted the
third word that I quoted as being an administrative request,
due to a rather generic regular expression. I can't really
explain it properly, since otherwise the same regular expression
would very possibly reject the message again.

On 2008-09-24 13:17, Scott Brim wrote:
> On 9/21/08 3:55 PM, Brian E Carpenter allegedly wrote:
>> ... the attachment to u n s c a l a b l e techniques such as PI addressing
> 
>> I think this is very do-able: PI for ten million larger sites,
>> with LISP-type mapping if needed, 
> 
> Could we de-conflate PI please?  We had to sort it out a couple of weeks
> ago.  PI is literally "provider-independent" -- just that.  It does not
> necessarily imply global routing or unscalability.  If you are talking
> about globally routed provider-independent blocks, please say so.

Actually, for the point I was trying to make, I don't think it matters.
As long as FIB sizes in the core routers are economically feasible,
we can stay as we are (with globally routed PI). When we reach the
economic tipping point for FIB size, future PI prefixes would need to be
mapped-and-encapped; I assume that tipping point would be somewhere
between a million and ten million prefixes. That doesn't avoid the
issue that transparent communication between current and "new" PI
prefixes will be required. Of course, that's what I was getting at
with my half-baked hierarchical proposal last November.

    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