7 ms·
IPv6: It's time to get on board
- pixl97 11y agoHey Suddenlink, Facebook is talking to you. Your neglected beta time is up, time for results.
- michaelmior 11y agoIs there any fundamental reason why IPv6 is faster than IPv4? My intuition is that the only reason things appear faster is because IPv6 communication is likely to occur using newer network devices (with newer strongly correlating with faster). Edit: The only technical difference I can see is that clients are expected to do MTU discovery which would probably result in better link utilization.
- pixl97 11y ago>IPv6 communication is likely to occur using newer network devices For cable modems it generally means moving to DOCSIS 3 supporting equipment, which yes, is generally faster.
- jewel 11y agoI don't think it's the fundamental reason but the checksum doesn't need to be recomputed at every stem, which used to be touted as a major advantage. (I imagine processing power at the router level is cheap enough now that this is no longer a concern, even for ten-gigabit links.) http://ipv6.com/articles/general/Top-10-Features-that-make-IPv6-greater-than-IPv4-Part4.htm http://ipv6.com/articles/general/Top-10-Features-that-make-I...
- rjsw 11y agoWith IPv6, you also don't have to compute the checksum twice at each end of the connection, with IPv4 you have one in the IP header as well as one for UDP or TCP.
- wmf 11y agoSome phones have native v6 but NATed (or even 464ed) v4, so the v4 traffic is going through a longer and slower path.
- syncsynchalt 11y agoRouter fragmentation is no longer allowed, that can be a cause (this is covered by your edit). One less checksum. On the meta level, core routing tables are much simpler but the use of CAM means that lookups are an O(1) operation, therefore any speed improvement would be due to the better scrutiny and route simplification that a smaller number of routes can afford. All this is offset with the v6 header being 40 bytes rather than v4's 20 bytes.
- api 11y agoThe big bottleneck is this: the vast majority of top-tier cloud providers (AWS, Azure, etc.) do not support IPv6 and seem completely uninterested in doing so. Until this changes I don't foresee a tipping point. I really don't understand why this is. None of these are old systems. IPv6 existed when AWS was built. Were they really designed from the ground up without IPv6 support? Of course I kind of know the answer. While most of hackerdom is pretty open to trying new things, most networking engineers are (in my experience) ultra ultra conservative and terrified of change. The majority of the network engineering community has dug in its heels like a stubborn mule at IPv6 and must be dragged grumbling into it.
- pixl97 11y ago>Were they really designed from the ground up without IPv6 support? Yup. It is simply amazing how many things I see devs push out that hard code badness (such as php ip2long, as an example). Intrusion detection that only will correctly resolve and send IPv4 to automatic firewall rules. Interfaces that truncate IPv6 addresses so information is lost. Who knows how many security problems are potentially added by this code that expects one thing and can get something else.
- api 11y agoA lot of it's just laziness. Perhaps we'd see more support if C included a built-in 128-bit type so IPv6 addresses could be shoved into a variable.
- kalessin 11y agoThat's what addrlen in all socket functions and struct sockaddr_storage are for?
- mprovost 11y agoYou've actually hit upon (part of) the answer right there. While IPv4 was designed around the (then) new 32 bit processors of the day, the committee that came up with v6 decided to ignore the underlying hardware reality. There's no corresponding type in C (or the underlying hardware, without going into SSE) that can store a v6 address natively. It just acts as a handbrake and makes it that much harder to implement. The question should be turned around, why didn't they design a protocol that could fit into a built-in type in C? After all, we've only just "run out" of v4 addresses about 20 years after it was first predicted to happen. Going to 64 bit would have kicked the can another few decades down the road, and could have led to a much faster implementation. You can't blame the "designers" of C (since there aren't any really, it's just the CPU manufacturers), it's the IETF that ignored the reality of hardware and programming languages.
- jewel 11y ago> knowledge of where IPv4 would fall short and its 2.0 version, IPv6 This is a comical example of the now common use of "2.0" to mean "new".
- Zikes 11y agoIt's a bit ironic since the "v6" literally stands for version 6. IPv4's 2.0 version would be IPv2, a step in the opposite direction.
- devit 11y agoNo, IPv4 + 2.0 = IPv6. Easy math.
- Zikes 11y agoBut if IPv4 was the first version then wouldn't the 2.0 be double that? IPv4 * 2.0 = IPv8 Why are we bothering with IPv6 when clearly IPv8 is the future?
- faitswulff 11y agoGoogle released V8 years ago, now it's all about VP9
- screaminghawk 11y agoWe are skipping v9 and going straight to v10 in an effort to show that this is a new product
- infogulch 11y agoI thought it was because legacy software detected the much older versions, VP95 and VP98, by doing a prefix search for "VP9.*" and they didn't want to break them.
- exabrial 11y agoFunny, the companies screaming the loudest about IPV6 are the ones that want to track users the most. I will be dragging my feat as much as possible on this.
- andrewpe 11y agoExplain how these companies will be able to track you better with IPv6. I would think it's the other way around since each device has (normally) a /64 possible addresses to use and routinely moves to a new address/holds multiple addresses
- devit 11y agoWith the default IPv6 addressing scheme the last 64-bit are the MAC address of your Ethernet card, which makes tracking trivial. Also, there is usually no NAT, meaning IP addresses are more likely to uniquely identify the user even without the MAC-based addressing. But of course setting a cookie allows better tracking, and Tor defeats tracking anyway for those wishing to intentionally defeat it, so in practice it probably doesn't make much of a difference.
- tangent128 11y agoThe "default" stateless auto-configuration uses the MAC for the "public" address, yes, but also has provisions for assigning the interface a temporary random address that can be used for outgoing connections. See RFC 3041
- ianburrell 11y agoPrivacy extensions, which use random, changing, local address instead of MAC, are enabled by default in most recent OSes. It means that individual devices can't be identified and can't be tracked across networks.mThe network prefix is assigned by ISP and has the same exposure as the public IPv4 address with NAT. The result is that privacy is the effectively the same as IPv4.
- scott_karana 11y ago
- dasil003 11y agoNot looking forward to having to support geolocation under IPv6.
- sadgit 11y agoI'm not sure why it would be that different.
- mentat 11y agoStill Google appears to get it wrong with whatever they're using on Chomecast. I've thought about trying to reverse where they're placing me for the weather information but it hasn't been worth it.
- dasil003 11y agoSize of the database is my concern.
- sadgit 11y agoThe database will get bigger but I'm sure it will accommodate ranges rather than individual IPS.
- sandstrom 11y agoE.g. Maxmind (commonly used for geoip lookups) support IPv6: https://www.maxmind.com/en/ipv6-information-and-faq https://www.maxmind.com/en/ipv6-information-and-faq
- porjo 11y agoThe author, Paul Saab, also presented on IPv6 at @Scale conf: https://www.youtube.com/watch?v=_7rcAIbvzVY https://www.youtube.com/watch?v=_7rcAIbvzVY