Making a protocol which supports 3.40282367 × 10^38 addresses just that teensy little bit more complicated…

by Brian Boyko
Editor, Network Performance Daily
The frustrating thing about IPv6 is that as much as we hate to think about it, as much as the transfer might irk and infuriate us – we’ve just about filled up IPv4, and that means that someday soon, we have to move to IPv6.
You can only delay the inevitable, which, right now, looks like it will happen in less than five years.
One of the big selling points of IPv6 was autoconfiguration. Sure, it’s a pain in the but to enter in longer numbers, so the IPv6 standard was designed so that you’d have to enter in those numbers less often. Through “stateless autoconfiguration,” you eliminated the need to set up DHCP.
But there are some advantages to the existing DHCP protocols, especially with visibility into the network, tracking and debugging features, as well as additional manual control – and so many companies, among them Cisco, are pushing for DHCPv6.
In the stateless autoconfiguration, you get true plug and play. The client is assigned an IP when it connects – without the need for a special server. Boom. Done. That’s perfect for portable devices, accessories like printers, and smaller operations in home and small office networks.
But DHCPv6 lets network manajers know what devices are connected to the network, as well as their IP addresses (and if necessary, to reassign them.) This is a major component in troubleshooting and monitoring the network to improve performance.
The downside: In addition to the complexities of changing over from IPv4, network administrators and engineers would have to manage a DHCPv6 network service as well.
Now, as mentioned, small home networks will find stateless autoconfiguration to be a boon, and ISPs will also be happy to avoid using DHCP to assign the addresses of the end-user. But for everyone else, DHCP is no longer a requirement – it’s a choice.
The odd thing about choices in corporate IT environments… it’s usually a short time to solve a problem, but it can take forever to make a decision.