+1 to Stewart's comment. Think growth. As one who argued for variable length address fields during the IPng debates - and time has shown the success gives us continual growth in MAC or IP addresses needed.
Since the IPng, we've figure out how to make short-cut addresses for routing (mpls, bier, etc.) - so the argument that the routers cannot handle it is no longer valid. Cheerily, Sue Hares -----Original Message----- From: Int-area [mailto:[email protected]] On Behalf Of Stewart Bryant Sent: Thursday, September 28, 2017 10:29 AM To: [email protected]; [email protected] Subject: Re: [Int-area] IPv-not-10. On 28/09/2017 14:23, JORDI PALET MARTINEZ wrote: > I don’t think even in those 500 years, we will run out of IPv6 > addresses, I hope you are right, but: I am sure that the original IPv4 designers thought that 2^32 was likely to be an infinity of IP addresses, but then the Internet was hugely successful, and now they are all allocated. I am also sure that the IEEE though that thought that 2^48 was likely to be an infinity of MAC addresses, but then VMs were invented, and now they are worried. People are proposing to use IP addresses for all sorts of things other than host addresses. For example, video fragments was one idea presented at the last IETF, but once that flood gate is opened, who knows what people will be wanting to address with IPv6 addresses, and how many they will need. So whilst 2^128 seems a pretty safe bet, history is not with us. - Stewart _______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area _______________________________________________ Int-area mailing list [email protected] https://www.ietf.org/mailman/listinfo/int-area
