We're already getting customers and inquiries today, but it isn't what I would call a barnburner, and we didn't expect it to be. We still believe it will be in the long term, but we're not looking to move $1 million per year in IPv6 products in the U.S.
For now, we're mostly working on acceptance, not by leveraging what we expect to happen here, but what we know from our IPv6 work in other parts of the world. Why did you decide to get into the IPv6 game here so early? It must be tough to profit from IPv6 at this point.
We chose to be there first because we wanted to make sure the U.S. had a viable platform provider for IPv6, and we already have to have those services available for our customers in Europe and Asia. As the technology advances elsewhere, U.S. companies might find themselves behind the curve. So if they're looking for a commercially viable provider, they can come to us. It seems there's interest in learning about IPv6, but little interest in deploying it, even from equipment vendors. Would you agree?
It depends whom you're talking about. I think Nokia and some other handset makers will say a lot about how IPv6 is better and other hardware makers will as well. That will be a watershed moment for infrastructure providers because, as a transport protocol, IPv6 can facilitate the transport of mobile data much better than IPv4 and any other proxy technology in use today. What's in it for device makers to carry the mantle for IPv6?
One of IPv6's big improvements [over IPv4] is in the area of mobility. IPv4 wasn't designed for mobility, but IPv6 was built from the ground up with mobility in mind and has those hooks in it. The IPv4 infrastructure today can't support the mobile infrastructure that handset makers are going to need. Do you see that as the primary driver of IPv6 adoption in the near term?
That's just a piece of the puzzle. It'll also be driven by other things that aren't considered part of the IP-enabled network today, like the IP-enabled icebox or other Internet appliances. Your readers may giggle, but it will be a very useful thing for the family.
For instance, it's expensive to have a security system installed in a house. But having devices like IP-enabled video cameras that record motion and send data over an IP connection could be very valuable, because data would go directly to you and not through a third-party service provider. Or a refrigerator may tell you when you're low on certain items and put them on your grocery list or order them from an online grocer. All these things can be done more easily with IPv6.
We very much think it will be. As an infrastructure provider, we can't be the key driver. We need partners like Panasonic to help us illustrate these capabilities. At the recent IPv6 Forum event, we teamed up with companies like Hitachi to demo capabilities like an IPv6 wireless camera that can take pictures of an auto accident, which can then be immediately sent to the insurance company to process the claim. If more of these everyday devices become IP-enabled, will that expose them to attack and create enormous security risks?
If it's secured with IPsec or SSL, it won't be an issue. For instance, with the IP-enabled home security camera, you could look to see if your kids were playing at home, but nobody else could. You'd typically log into an SSL server today, but with IPv6 you'll be able to use IPsec, which makes it even simpler to use because you're putting security in a different part of the stack. To date, the key driver for enterprises considering IPv6 has been the potential to make network management easier. Will that potential be realized?
I think so. There are two examples I like to cite. IPv6 does allow for LAN-based device auto-configuration. So you can plug a Windows XP box into an IPv6 network and it'll auto-configure itself. In IPv4, you have to configure a DHCP server and ensure the additional address pool is available. So for the cost of a $1,000 node, you could deploy any number of systems overnight simply by having people plug them in. It's not a huge time savings, but if you're saving a few minutes per box, then you're saving a significant amount of time.
Also, IPv6 enables you to put more of the "smarts" in the network to make management easier. IPv4 has problems with subnetting efficiently and dealing with routing packets efficiently. The IPv6 routing implementation offered by Cisco Systems makes that much easier, so you don't have to sit down and do subnet calculations. Organizations like the University of New Hampshire's InterOperability Lab have been working to make the IPv6 transition easier for enterprises, but will it still be difficult?
Yes, it will be somewhat, but fortunately those groups are going through most of the pain for us. We're able to leverage a lot of the stuff those guys have done. But customers won't have to go through all the painful changes we endured when we were improving IPv4.