On Wednesday, March 13, 2013 at 1:45, Markku Miettinen wrote: (I'd saved a few mails from this thread where I wanted to point out something) > But in principal amounts of users and servers with native IPv6 probably > follow the same trend in the long run?
Yes, quite likely. On average. I actually think, on average, the servers will be ahead. If we just assigned a random server from somewhere on the globe to each client, IPv6 would be fine. But we're used to doing better than that with IPv4. As I've said before, I don't want people to get a worse service just because they enabled IPv6. The problem is not so much the number of IPv6 servers relative to users but the poor distribution (most countries have basically none). The absolute number in most places is too low to make the load balancing stuff we have work well. I haven't looked but I also suspect that the diversity of network and hosting is much poorer for IPv6 than it is for IPv4 (lots of servers with a few IPv6 enabled rented server places, that sort of thing). Ask -- Ask Bjørn Hansen - http://askask.com/ _______________________________________________ pool mailing list [email protected] http://lists.ntp.org/listinfo/pool
