[6bone] IPv6 routing table size

[Bill Manning]

	a single "million" is either way too large or 
	way too small.  Depends on if you beleive the
	aggregation model will hold. A prgmatic design
	choice would be for supporting a table size of
	2x32nd number of entries, presuming CIDR.
	Is such a design practical? Not with todays
	technolgies.  How much social engineering
	can be done to ensure that choice is not
	restricted?


% > Jeff Thomas wrote:
% > was building an IPv6 router with over a million
% > entries in their routing table.  Am I off track?
% 
% A _million_ ? whoa. I count 245 entries in the global v6 table now, and
% I am not aware of any v6 ISP with more than a few dozen thousand
% customers.
% 
% > Large enterprises today may have tens of thousands of routes
% > in v4, but using v6 wouldn't they be an order of magnitude
% > smaller?
% 
% One million subnets, yes; this would be totally unaggregated though.
% Should not be more than 50k-80k routes aggregated.
% 
% I am not sure I would agree with an order of magnitude smaller, but I
% think that a v6 routing table should be smaller than the v4 equivalent.
% The biggest (v4) routing table I have seen myself is 200k and change.
% 
% Michel.
% 
% _______________________________________________
% 6bone mailing list
% 6bone@mailman.isi.edu
% http://mailman.isi.edu/mailman/listinfo/6bone
% 


-- 
--bill