[6bone] Re: IPv6-only IXP's are absolutely wrong

[Bill Manning]

% > 	fabric "optimizations" - framing support, buffer sizing, MTU, etc.
% > 	a decent L2 fabric will be able to accomodate the larger MTUs
% > 	of native v6 and won't complain about divergent framing.
% > 	"Smart" fabrics tend to be tuned to IPv4.
% > 
% > 	then there are mgmt issues. most L2 fabrics do not have up to date
% > 	v6 mib support, so the stats/traffic collection is not as accurate
% > 	as it should be.
% > 
% I believe most of your points would be satisfied by "run different vlan's
% for v4 and v6'.  Seems sane to me, and works quite fine for L2 exchanges.  
% If you need L3 for some reason the issue of HW/SW forwarding may become
% important.
% 
% On smaller exchanges, building dedicated peer-to-peer VLAN's is IMO also
% an option worth considering.
% 

	presuming vlan support on the fabric
	presuming accurate auditing/monitoring capabilities for v6
	presuming a switch that won't eat your lunch (performance
	hits) with larger packets, different framing, MTU than "normal"
	IPv4.

	-but- given the level of v6 native traffic, esp. in relationship
	to the level of v4 traffic, the arguments may be moot.

	These were simply points as to why one might consider 
	running v6 only exchanges.  I think that the issues are more
	pronounced with transit providers infrastructrure, which, (IMHO)
	is one reason that v6 is having a more gradual rampup than 
	some would have liked to see.


-- 
--bill