Slashdot Mirror
The Impending IP Crisis
Factomatic writes "With the supply of IP addresses expected to run out by 2005 due to the popularity explosion of the Internet and the expectation that everything from your phone to your washing machine will soon have its own IP address, Alex Lightman, CEO of Charmed Technology and chairman of last month's North American IPv6 Global Summit tells the New York Times "we're going to need something like 100 IP addresses for each human being." IPv6 will increase the supply of addresses from 4 billion today to a number in excess of 35 trillion that is "so big that there's not a word for the number," says Cody Christman, director of product engineering for Verio, which offers IPv6 in San Francisco, Washington and elsewhere. The article is a good layman's backgrounder on the looming IP crisis."
Who needs a new word to describe the number of possible addresses? It's just 1/2.9387358770557187699218413430556e+61st of a google.
Stop by my site where I write about ERP systems & more
To quote the article "Such sensors could allow people to operate devices from anywhere there is an Internet connection." and "Now that the address space is available, the next step is figuring out how to use it."
I've got an idea, a internet connected toilet. "Using a cellphone in Los Angeles", I could flush the toilet at my home remotely and have the toilet seat drop down automatically (you know, to keep domestic tranquility). I could even call the toilet to see if anyone is using it.
I better go patent it...
Accentuate the positive, don't waste your mod points on the negative.
IPv6 will increase the supply of addresses from 4 billion today to a number in excess of 35 trillion that is "so big that there's not a word for the number,"
how about "thirty six trillion" ?
unless a new prognostication that 'the end is nigh, in 2005' passes as news. everyone knows it's gonna happen. just as we all know that with NAT and proxies, most of it can be safely delayed by tech companies until they have an outside fiscal force to upgrade.
and i doubt my fridge will have an IP address anytime -before- ipv6 starts to be rolled out en masse.
as with all pure tech - it needs that killer app. something needs to come out that is so fantastically great that everyone has to have it - and it needs to require ipv6. until then - at best we'll be going dual-mode.
good luck finding that app, and educating users what it is, and what it does.
// "Can't clowns and pirates just -try- to get along?"
Not too hard.
Backbones should switch over first, proxying ipv4 over ipv6, then propogate downwards.
When it hits users, they'll have an ultimatum. Upgrade within the next 180 days, or j00 are fux0red.
As for the OS and device makers, simply make dhcp check ipv6 first, then fallback to ipv4. That'll be transparent for all the chuckleheads who would ignore the "switch" thing.
-
ping -f 255.255.255.255 # if only
I want to see IP as more of a general resource like electricity or water. You just plug anything into your wires/pipes, and it gets full access to the resource. Want more things getting water such as a washing machine? Then just run another pipe to it and it's got access. The current hacks of NAT are equivalent to only being allowed to install one tap in your house, and "proxying" the rest with buckets. Why cant it be like a water or electricity supply?
Those saying 'we have plenty of space left' obviously dont realise that the reason for this is that the current allocation policies for IPv4 make it impossible to get space for arbitrary devices. Yes, if you only allocate one IP address per gateway, of course you wont run out for a while. But that then mandates the use of ugly hacks such as NAT. A single tap per house/organisation.
To make full use of the potential of the net, one must be able to freely allocate IP addresses to any devices that want them, no matter how trivial it may seem today. Back when IP was invented, it was never in anyones wildest dreams that there would be an address shortage. There were barely a hundred hosts yet 32 bits of space. Look at what's happened in 20-odd years!
Lets not make the same mistake today.
Sparks:Gadget:Beer Maker
IPv6 is bad because Cisco routers suck. No, wait, "Many of Cisco's routers" suck. You can' be serious! Once IPv6 gets off the ground, IPv6 will become fast path and eventually IPv4 will be dropped to legacy mode.
... most of the internet protocols are very wasteful. On the other hand, they are easily debuggable with relatively simple tools. This is a trade-off, obviously, and IPv6's choice is not per se good or bad, it's just different. We will see whether it will have a significant overhead. I say getting rid of spam is a better way to reduce bandwidth requirements on the internet than talking about header sizes.
About your point 2: IPv6 does not actually give out all those 2^128 IPs. The first half is for the network part, the second 64 bits are for the host part. This is necessary because autoconfiguration (which is really great, by the way!) uses a 64-bit part. The IPv6 autoconfiguration is stateless, by the way, which means it will also work without a DHCP server and it won't need reboot if the routers were down when the autoconfiguration process started.
The point about having this many addresses is that you never ever want to have to come into the remote possibility to have to switch to IPv8 because IPv6 is too small. And when you rant about the IPv6 header being 20 bytes larger than the IPv4 header, consider that the overhead of the TCP header (20+ bytes), the HTTP header (300 bytes), the Email header (500 bytes?),
IPv6 is ready for prime time. People are using it (I, for example). You can buy access to IPv6-native backbones. All the major OSses support it. There is really no excuse not to be already using it.
Because your other devices will want to keep their identity even when not at home. Imagine having an IP-based telephone as a (slightly contrived) example. You want to be able to route to it no matter what network it resides on at the moment.
I use my laptop in a number of places; home and at the department is the most common places but also others. Moving from place to place is a bit of a pain, though - I need to get a new IP address, change the SMTP server and so on, and setting up other stuff so I am allowed to access it no matter where I am is painful and error prone. If my laptop could keep its identity irrepsectively of where it is physically located on the net it would simplify life a whole lot for me.
NAT works pretty well for the stuff we do today, but it precludes a lot of interesting uses, and is actually quite painful compared to the possible alternative.
Trust the Computer. The Computer is your friend.
No, we should withdraw all the A-class networks that are unnecessarily allocated to US companies.
OTH - I'd rather move to IPv6.
it's in my head
Backbones are already upgrading to IPv6 enabled software and hardware. My employer has plans to run dual-stack IPv4 IPv6 later this year which means that any existing IPv4 customer can give us a call saying "enable v6" and we can do it that day. (assuming they have their hardware/software in place). No tunneling, no 6to4 gateways, it'll just work. I see no long-term viability of the 6to4 gateways, in the same way that we didn't see caches go mainstream for every internet user. (yeah yeah, some of you will claim bittorrent is a large distributed cache, and while that might be the case, i'm talking about for most of the general public, the AOL/IE users that don't know how to spell IP).
If you also see one of my previous comments on IPv6 here about who is supporting it (note, what you might define as a backbone isn't what the rest of the network might..) and has existing routes in the tables, you'll get an idea of who is at least prepared for the new future of impossible to read ip addresses.
If everyone runs dual-stack v4v6, you'll see the ability to access your existing services while continuing to be able to gain access to the IPv6 content. Personally, I've seen that in cases like where a RedHat release comes out, I can get faster transfer rates going to the IPv6 mirror than the IPv4 mirror. Everyone is hammering the v4, which makes the v6 available for me :). I'm just waiting for Linksys (now cisco) and the other consumer product people to realize that they need to upgrade their devices so they can do IPv6 nat for those cablemodem routers, etc..
Here's where I think that the local loop (dsl, cable) providers can go and start to seriously make money and make IPv6 viable: IPv6 enable your network, then offer VoIP services over SIP enabled devices. This way you don't run out of numbering space (ip and pstn). (Trivia: how many ips would it take to convert the existing PSTN network to VoIP, if each phone number required an IP address).
Which is: 340 undecillion, 282 decillion, 366 nonillion, 920 octillion, 938 septillion, 463 sextillion, 463 quintillion, 374 quadrillion, 607 trillion, 431 billion, 768 million, 211 thousand, 456.
A far cry from "35 trillion". To give you an order to this magnitude, some Australian scientists recently announced that there are 70 sextillion stars (give or take) in the known universe.
It may be pedantic, but someone who is so blinded by their work that they make hysterical claims that there's no word for the number they're pushing doesn't make me want to buy into their idea so quickly.
1. Cisco routers suck at IPv6.
Cisco is working on hardware support for IPv6 for most platforms. As it becomes more widespread, they will develop full hardware routing for all devices. It's a chicken and egg problem, but a lot of people have decided to start somewhere.From what I've seen, there has been more interest in getting IPv6 running in the last 6 months than there ever has been.
2. There are too many addresses.
NAT doesnt work when you have devices that need to be addresses externally. What NAT-Portfw port is your device 'X' listening on? Hmm?? IPv6 is designed such that address space is _not_ a resource, but a method of being able to access _any_ device attached to the net. I dont _want_ to have to explain to someone why I want an address. I just want the thing to work. /24, /25, /26, etc. Yuck! With Ipv6, you never have to worry - there's enough space in _one_ subnet to scale indefintely. No more subnet-resizing games.
Another benefit is address management. With Ipv6, the days of deciding how big your subnets are will be numbered. Every subnet has trillions of addresses. Nowadays you have to wonder whether your subnet needs a
3. IPv6 addresses are too large.
Ipv6 is designed to be very hierarchial. The top organisations get /32s, the next one down get smaller subnets (/48), etc. The routing table will no longer be populate with tiny piddly subnets (eg /24s today). Go into an aggregate and things will behave.
And as for the routing table size, modern routers have oodles of memory, on average 512Mb RAM. A full BGP table of 130k routes takes up around 64Mb of that at most. IPv6 will have better aggregation, so smaller numbers of routes, and my the time it gets large, standard memory on routers will not have a problem storing the table.
4. The IPv6 header is too large.
3.4% longer if you use a 576 MTU. Use a sensible 1500 byte MTU and your downloads will not be much slower at all. Anyway, the elimination of fragmenting, the simplification of subnetting, and the many other benefits will far outweigh your 20-byte concerns. See my other post on security too (no more network host scans).
IPv6 is not ready to fully replace IPv4 overnight now, sure. But it's gaining a heck of a lot of momentum and by the time your concerns will actually become enough to be worried about, they will have been either solved or rendered moot.
Another nice thing about IPv6 (at least now), is that it's a return to the good ol' days of the net when everyone was friendly with each other. 15 years ago you askes someone for addresses or transit and you were quite likely to get it for next to nothing. Ask for addresses or transit/peering with IPv6 today - you're likely to get it, with a friendly response. it's a great community.
Sparks:Gadget:Beer Maker
Of course we'll run out of IPv6 addresses.
... but surely something will be invented that calls for more addresses.
Not right away
For example, teleportation might require separate addressing for all possible energy states of all elementary particles in the teleported object.
Don't say it can't happen. Remember when 64k was all the memory anyone would ever need? and a megabyte hard drive was out of your price range?
-kgj
Pardon me. The ACTUAL number is:
Three hundred forty undecillion two hundred eight-two decillion three hundred sixty-six nonillion nine hundred twenty octillion nine hundred thirty-eight septillion four hundred sixty-three sextillion four hundred sixty-three quintillion three hundred seventy-four quadrillion six hundred seven trillion four hundred thirty-one billion seven hundred sixty-eight million two hundred eleven thousand four hundred fifty-six.
Or just: 340 282 366 920 938 463 463 374 607 431 768 211 456