On Tue, Feb 11, 2003 at 09:25:24AM +0100, Kurt Erik Lindqvist wrote: > > 1) Address space shortage > 2a) No scalable PI solution > 2b) No scalable IDR solution > > I said it before and I will say it again, without a solution to 2a, no > enterprises will go to IPv6, with no enterprises on IPv6 there are no > revenues for the ISPs in IPv6, with no revenue from IPv6 the ISPs will > not go to IPv6. Just because I can reach a webpage over IPv6 doesn't > make the web-page more interesting. If it isn't more interesting I > can't charge extra for it. Going to IPv6 will cost the ISPs. Sorry. It > doesn't work out. We need to progress on multi6 if this is to take off.
Yes, but not for all networks. I appreciate many people don't view the academic research networks as commercial networks, but there is a large potential IPv6 user base in that community for whom address stability has been available through pre-CIDR allocations. In the UK example, the 200 or so universities will go from 160 IPv4 (PI) prefixes to 1 single IPv6 (PA) prefix, but the provider being the non-commercial "independent" NREN makes the address space in effect PI (JANET has a /32 allocation). Each of the 25-30 European NRENs are likely to be in a similar position, so you could see 25-30 SubTLA's replacing (well, living alongside :) around 3,000 IPv4 prefixes. Behind those 25-30 SubTLAs are maybe 30 million staff and students at the universities. Well, that's one view. Many regional networks may seek to multihome. Some universities might seek SubTLA prefixes (many obtained pTLA space, and some are LIR's for the benefit of IPv4 multihoming, but multihoming to universities isn't common because they're tied in, for better or worse, richer or poorer, to the NREN service). I think multihoming is important in the shorter term for a number of classes of networks, including some large enterprises. IPv6 multihoming for home networks is further off, and is the point at which you would be looking at the "1 billion multihomers" scenario. But for some (many?) users/networks, multihoming isn't critical to adopt IPv6. Tim -------------------------------------------------------------------- IETF IPng Working Group Mailing List IPng Home Page: http://playground.sun.com/ipng FTP archive: ftp://playground.sun.com/pub/ipng Direct all administrative requests to [EMAIL PROTECTED] --------------------------------------------------------------------
