3 ms·
Can anyone tell me(or point to a good blog post) why IPv6 is more than simply IPv4 with 128 bits for an address vs 32?
by oh_sigh 6y ago
Can anyone tell me(or point to a good blog post) why IPv6 is more than simply IPv4 with 128 bits for an address vs 32?
- freeone3000 6y agoMain differences: IPv6 uses multicast over unicast/broadcast. ARP is handled at the IP level rather than the ethernet level, and is done via solicited multicast. Anycast exists, so OSPF is essentially part of the protocol. (Theoretically, you could dispatch at random, but you probably already built an OSPF tree for v4 routing) The addresses are strictly heirarchial, with variable-length hostmasks -- this is even bigger than CIDR, as it's assumed subnets are actually masked off of a larger subnet (There's also some boring stuff like the checksum being dropped) These are all features of how IPv4 networks worked in 2004, but none of them were at the IP level.
- fulafel 6y agohttp://www.faqs.org/rfcs/rfc1678.html http://www.faqs.org/rfcs/rfc1678.html IPng Requirements of Large Corporate Networks http://www.faqs.org/rfcs/rfc1686.html http://www.faqs.org/rfcs/rfc1686.html IPng Requirements: A Cable Television Industry Viewpoint Short answer: there was more stuff people wanted to fix.
- arpa 6y agoSecond system effect.