IPv4, has been around since 1981 and is quite an engineering achievement to have remained in use for so long. Unfortunately though, IPv4 has become a victim of its own success. The initial design of IPv4 never took into account that the Internet would become so popular. Let's examine why IPv6 is lurking in the wings hoping to overtake IPv4 for Internet supremacy. IPv4's drawbacks
IPv4 is becoming quickly outdated because of the way that it designates network IDs. Internet backbone routers contain routing tables with over 85,000 routes. These routes are combinations of flat and hierarchical routes. While the current Internet backbone routing tables are getting the job done, they're inefficient. Another reason for moving past IPv4 is that TCP/IP can be tough to configure. Before you e-mail me over that statement, let me clarify. Certainly, most network professionals can configure TCP/IP with their eyes closed. However, the first time you manually configure TCP/IP, it can be a little intimidating. A lot of people want an IP protocol that doesn't rely on a DHCP server or complex manual configurations. Security has also become an issue for IPv4. Everyone is concerned about the encryption of data flowing across the Internet. There are lots of ways of encrypting IPv4 traffic, such as using the IPSec protocol. Unfortunately, all of the IPv4 encryption methods are proprietary and no real standard encryption methods exist, although some encryption techniques are more widely used than others. A final challenge has been the real-time delivery of multimedia content and the necessary bandwidth allocation that goes along with it. A bandwidth allocation method called Quality of Service (QoS) has been used with IPv4. While QoS does work, there are a number of different interpretations of the IPv4 QoS standards. This means that not all QoS-compliant devices are compatible with one another. The IPv6 protocol
IPv6 has a number of new features designed to address the shortcomings of IPv4, including a new IP header format, a larger address space, a more efficient routing infrastructure, stateless and stateful address configurations, enhanced security, and standardised QoS support. The new header format
The first notable feature of the IPv6 protocol is a newly designed IP header. It's designed to make the protocol more efficient by keeping overhead to a minimum. An IP packet header is made up of required components and optional components; in IPv6, the required components are moved to the front of the header. Optional components are moved to an extension header. This means that if optional components aren't used, the extension headers aren't necessary, reducing the packet size. The downside to the new header is that it isn't compatible with IPv4. If a router is to handle both IPv4 and IPv6, it must be configured to recognise both protocols. You can't just set up a router to recognise IPv6 and expect it to be backward-compatible with IPv4. Larger address space
Perhaps the most compelling reason for moving to IPv6 is the supposed shortage of IP addresses. IPv6 uses 128-bit source and destination addresses. There are theoretically over 3.4x10^38 possible addresses that can be addressed by the IPv6 protocol. Furthermore, this new structure allows for more levels of subnetting than are available with IPv4. Some people speculate that because of the large number of addresses that IPv6 allows, NAT technology may soon become a thing of the past. More efficient routing
The Internet is hierarchical in nature, and the IPv6 protocol is designed with this in mind. Think about it. The computer you're using right now doesn't have a direct connection to an Internet backbone. Instead, you're probably behind a NAT firewall, which is connected to an ISP. That ISP may be connected to another ISP or to a backbone router. Either way, a packet must make quite a few hops to go from an Internet backbone router to you. The IPv6 protocol is designed so that Internet backbone routers will have much smaller routing tables than they have now. Instead of knowing every possible route, the routing tables will include routes to only those routers connected directly to them. The IPv6 protocol will contain the rest of the information necessary for a packet to reach its destination. New configuration options
One of the coolest things about IPv6 is the way it's configured. While you can still manually configure IPv6, or lease an address from a DHCP server, there is a new automatic configuration option available. If an unconfigured PC tries to connect to a network that doesn't offer a DHCP server, the PC can look at either the network's router or the other PCs on the network and determine an address that would be appropriate for it to use. This technique is referred to as link local addressing. Integrated security
IPSec is available in some implementations of IPv4, but it's completely integrated into IPv6. Any computer that's running IPv6 will support IPSec encryption, regardless of the computer's operating system. Standardised QoS support
IPv6 also includes standardised support for QoS. The QoS implementation is set up so that routers can identify packets belonging to an individual QoS flow. This allows those routers to allocate the necessary amount of bandwidth to those packets. Furthermore, QoS instructions are included in the IPv6 packet header. This means that the packet body can be encrypted, but QoS will still function because the header portion containing the QoS instructions is not encrypted. This will make it possible to send streaming audio and video over the Internet with IPSec encryption, but in a manner that guarantees adequate bandwidth for real-time playback. Get ready to move
IPv6 is a huge improvement over IPv4. You should expect many ISPs to start supporting both IPv4 and IPv6 as more demand is made for IPv6. The IPv4 protocol will be gradually phased out, but this may take some time. For a weekly round-up of the enterprise IT news, sign up for the Enterprise newsletter. Tell us what you think in the Enterprise Mailroom.