IPv6 Rollout Japan, China in 2005

Killjoy_NL writes "The digitimes have a piece that is reporting that IPv6 will be rolled out in China and Japan in 2005. Makes me wonder when the rest of the world will follow suit" We had a good piece a couple months back about the state of IPv6. CowboyNeal is ready!

Comments

Re:How many addresses?

[Sediyama]

IPv6 has an address space of 128 insteade of 32 from IPv4. IPv4: 256^4 IPv6: 65536^8 Exemple IP would be: FEDC:BA98:7654:3210:FEDC:BA98:7654:3210 1080:0:0:0:8:800:200C:4171 3ffe:2a00:100:7031::1 1080::8:800:200C:417A ::192.9.5.5 ::FFFF:129.144.52.38 2010:836B:4179::836B:4179

Re:IPv6: Not Ready For Prime Time

[Peaker]

1. Cisco routers suck at IPv6. Many of cisco's routers use the router's CPU to process IPv6 packets instead of the fast-path. The reasons for this are explained in the next few points. While Juniper's routers are substantially better at IPv6 than cisco's, IT managers are often restrained by insane corporate policy that dictactes the use of cisco.

Oh, you're right, Cisco doesn't support IPv6 well, lets just drop the whole thing. What a great point you have here.

2. There are too many addresses. There are 16.7 million addresses per square metre of the earth's surface, including the oceans. This is overkill. The world does not need more than the 4 billion addresses available with IPv4, and I challenge you to come up with an application that requires that many. Assuming that you can actually come up with one, it could easily be solved with Network Address Translation, or NAT as it is commonly known.

Oh, ofcourse having too-many addresses in itself is a huge problem (we might have to leave some addresses unallocated - the horror!). Yet another great point.

3. IPv6 addresses are too large. An IPv6 address is 128 bits in size - 64 bits of which are reserved for addressing hosts, and 64 bits of which are reserved for routing. One thing that is cool with IPv6 is address autoconfiguration. Take your 56-bit MAC address on your ethernet card, ask for 64-bits of network prefix, bang it together with EUI-64 and you are set. The problem with a 64-bit network prefix is that routing tables become massive. Just do the math and you'll see that extreme amounts of memory are required to hold routing tables.

If you're trying to say 128-bit is too large because routing tables become too large, that's simply ignorant. IPv4 addresses are so small that they cannot easily be geographically/connection-wise allocated. This means that routing tables became large because of the complexity of IPv4 addressing.
128-bit allow much simpler addressing schemes which will actually make routing tables much simpler, and probably smaller, even though each address is a few bytes longer.

4. The IPv6 header is too large. An IPv4 header compact at 20 bytes in length, while the IPv6 is bloated at 40 bytes. That's right people, each one of your IP packets has twice as much overhead as before. While this may not sound much, IP networks have a requirement that the minimum MTU supported must be 576 bytes. That means that where you might have got 556 bytes of data in your IP packets, you now get 536 bytes. This means that downloading stuff will take 3.4% longer.

A) The fact that the minimum MTU required is 576 bytes is meaningless, real MTU's are much higher.
B) Few networks actually use all of the potential ether/link bandwidth all of the time, so a few percents of extra/lower usage don't matter much.
C) The overhead of IP packets is almost negligible anyway, and this does not change it considerably.

You are obviously a troll.

Re:IPv6: Not Ready For Prime Time

[Zarhan]

Okay, since there seem to be folks that are actually taking this guy seriously, I guess I really have to debunk him point by point:

1. Cisco routers suck at IPv6.

One word: IOS 12.3.
Also, in 12.3T series, you get functionalities like stateful firewalls for IPv6. Check out Cisco's IPv6 status here.

2. There are too many addresses.

640 kB should be enough for everybody.

And yeah, I know Bill Gates never said that.

IPv6 addresses are too large. The problem with a 64-bit network prefix is that routing tables become massive.

This has been addressed. Summary routes are there. The IPv6 addressing structure is quite hierarchical, so even that /64 prefix that the end-user sees is quite nicely broken down into categories.

4. The IPv6 header is too large

Minimum MTU for IPv6 is 1280 bytes, not 576. Also, IPv6 header structure is extendable, ie the last field in IPv6 header is a pointer to an optional field. This optional field can in addition to it's own information refer to even more fields, in daisy-chain fashion. There is much bloat in IPv4 headers and lots of bit-alignment problems when building hardware to forward IPv4. IPv6 addresses these details by daisy-chaining optional headers and keeping the stationary fields simple.

For starters...

[HopeOS]

This is the exact same comment he posted to the last article about IPv6. That comment was modded 0, Troll. Check his post history at January 11th. Also notice that in the last 15 posts, three of them are sitting at -1 and another two are moderated troll.

If you want a decent rebuttal of his silly argument, just go back to his previous post.

Re:Dont' worry: We (the US) won't.

[Monkey]

Every Cisco router we have that will run IOS 12.2(2)T is capable of IPv6. This even includes most of the lowly 2500 series routers that we bought in '95.

Re:IPv6: Not Ready For Prime Time

[temojen]

Sounds like a Troll, and I could just mod him down, but I feel more like argueing back.

1. Cisco routers suck at IPv6. Current models, perhaps. But cisco is not the only router manufacturer, and routers can easily be made from a commodity PC with 2 (or more) network interfaces and commercial or open-source software (Linux/OpenBSD/Solaris or whatever).
2. There are too many addresses. How can there be too many addresses? There may be more than is currently forseen as needed, but I don't see how there can be too many. NAT is no solution as it breaks a great many protocols including H.323. As IPv6 is intended to make it possible for devices like telephones to live on IP networks, not breaking H.323 is mandatory.
3. IPv6 addresses are too large. Bits are cheap, future re-specifying of protocols is not. A routing table does not need to have an entry for every possible address, just what ranges are reachable from each interface.
4. The IPv6 header is too large. MTU stands for Maximun Transfer Unit so with a physical layer with the smallest frame size will have 3.6% more overhead. These networks tend to be short-haul connections in noisy environments. Longer-haul networks (and/or) with less noise tend to have MTUs closer to 10,000, so with these networks you have closer to 0.2% more overhead. This doesn't seem like such a big deal to me.

Re:Slashdot IPv6

[Joe U]

Last time I checked (late last year) Slashdot didn't do IPv6

Last time I checked, Slashdot didn't do valid HTML either.

Re:Other Countries slow to follow

[espo812]

Japan also rolled out 3G wireless before everyone else. Have we incorporated as much as Japan? [...] But only because they don't have all of the infrastructure flaws other countries do.

Japan is also much more densly populated, so it's more cost effective to roll out advanced wireless products there. If by "infrastructure flaws" you mean the ability to make an upgrade in a smaller area to benefit more customers, then yes Japan has fewer of those. They also have a state run telephone network that is increadibly expensive - not exactly a desirable infrastructure.

In Europe IPv6 is happening also

[jordip]

See this document and this site. And this one for the rest of the world.

Also news posted at the IPv6 Cluster.

By the way, a new tunnel broker is available here, also with Spanish instructions at 6SOS.

Re:China?

[Anonymous Coward]

except that a list was made and the US *IS* the no. 1 spammer. it was on /. a month ago or so.

Don't get a tunnel!

[Wesley Felter]

As we have to point out in every IPv6 thread, 6to4 is more efficient than tunnels.

Re:China?

[mark-t]

In actuality, each "customer" would be an entire ISP, corporation, or institution, not just one particular individual. ISP's could assign blocks of IP's to individual domestic subscribers as they see fit (I can't imagine that a domestic subscriber would need more than a few thousand IP's, and that's allowing for every device in his household, from his VCR and washing machine to his toaster and alarm clock radio, was connected to the internet with a unique IP, which I don't think would ever actually happen in practice (local subnets with private IP's behind a NAT are fine for certain types of restricted usage devices anyways).