Gearing up for IPv6

IPv4 has its limitations, so IPv6 is the future, says Shrikant Shitole, Business Development Manager, Cisco Systems India & Saarc.

Picture this. You are not able to log on to the Internet because your computer says, “Sorry, all IP addresses are now being used!” What does one do in circumstances like these? Right now nothing. With the rapid growth of the Internet and the proliferation of wireless devices that require unique IP addresses, IPv4 is beginning to show its age.

But the rumblings of the next-generation IP protocol—version 6—are growing louder. IPv6, or Internet Protocol version 6, is the ‘next generation’ protocol designed by the Internet Engineering Task Force (IETF) to replace the current version of the Internet protocol, IP version 4 (‘IPv4’), and is now said to be the remedy for shortages of IP addresses.

The case for IPv6

The escalating demand for IP addresses is the catalyst for IPv6. It’s estimated that, in the wireless arena alone, more than a billion mobile phones, personal digital assistants (PDAs), and other wireless devices will require Internet access, and each will need its own unique IP address. Additionally, billions of new, always-on Internet appliances for the home—from the TV to the refrigerator to the utility meter—will also be connected through various technologies. Each of these devices will also require its own IP address. Clearly then, with the exponentially increasing demand for IP addresses, the world is outgrowing IPv4.

The need for more addresses is compounded by the fact that addresses have not been distributed evenly worldwide. Although IPv4 theoretically can support as many as four billion unique addresses, the actual allocation of space has locked up nearly 75 percent of these addresses. In contrast to IPv4, which has 32 bits of address space, IPv6 has 128 bits of address space, pushing the theoretical limit of unique IPv6 nodes to roughly 3.4 x 1038 or about 340 billion billion billion billion unique addresses.

A look around the world shows that IPv6 adoption is taking place. Japan was the first country to put in place a national strategy for adoption of IPv6 in 2000. In the US, North America IPv6 Task Force was set up in December 2001, and it has been promoting the adoption of this technology. Even China announced a national strategy in 2004 for promotion and adoption of IPv6. In India, the industry has just recently realised the need. TRAI issued its recommendations on ‘Transition from IPv4 to IPv6 in India’ in January 2006.

The catalyst in action

IPv6, as specified by the IETF, offers the following benefits to enterprises:

• Expanded IP addresses to accommodate the widespread proliferation of Internet devices such as personal computers, personal digital assistants, wireless devices and new Internet appliances.
• IPv6’s auto configuration feature or auto-discovery will eliminate today’s labour-intensive and expensive approach of dynamic host control protocol (DHCP) servers that most large organisations use to manage their IPv4 addresses.
• IPv6 will use a ‘stateless autoconfiguration’ that combines an interface ID number, such as the machine’s existing MAC address, and a network prefix from the local router, to assign its IP address instead of one allocated by a DHCP server.
• IPv6 will also provide a peer-to-peer applications architecture that will support a global, end-to-end addressing scheme. This set-up will eliminate the need for network address translation devices at the edge of large enterprise networks; these can slow down the encryption process and are inefficient for applications such as voice over IP, mobile IP, and distributed gaming.

Apart from the obvious benefits to large enterprises, this represents a chance for aggressive and forward-thinking ISPs to develop an IPv6 network to carry gaming and eventually other types of digital content such as movies, television programming and music. Other than the new revenue streams, IPv6 should also bring about cost-reduction in larger networks. With the improvement in routing, inherent security, and the auto-configuration capability, tomorrow’s networks will cost less to deploy, maintain and operate. Once you add services that either reduce expenses elsewhere, i.e. communications (VoIP, IP videoconferencing, etc), or improve productivity (true mobile access to files, information, inventory, etc), the cumulative ROI effect is positive for both adopters and developers.

The question isn’t ‘if’ but ‘when’ the world will migrate to IPv6. ISPs and enterprises will have to move hand-in-hand towards the adoption because it involves expenditure for both. Each will need to nudge the other forward in the adoption process.

Deployment techniques

Network designers recommend deploying IPv6 first at the edge and then moving towards the network core to reduce the cost and operational impact of integration. The key strategies used in deploying IPv6 at the edge of a network involve carrying IPv6 traffic over an IPv4 network, and allowing isolated IPv6 domains to communicate with each other before making a complete transition to a native IPv6 backbone. It is also possible to run IPv4 and IPv6 throughout the network, from all edges through the core, or to translate between IPv4 and IPv6 to allow hosts communicating in one protocol to communicate transparently with hosts running the other.

All these techniques permit networks to be upgraded and IPv6 deployed incrementally, with little to no disruption of IPv4 services. The four key strategies for deploying IPv6 are:

• Deploying IPv6 over dual-stack backbones. This technique allows IPv4 and IPv6 applications to co-exist in a dual IP layer routing backbone. All routers or a portion of them (for example, access CPE routers and aggregation routers are dual-stack, but core routers stay as they are) in the network need to be upgraded to be dual-stack, with IPv4 communication using the IPv4 protocol stack and IPv6 communication using the IPv6 stack.
• Deploying IPv6 over IPv4 tunnels. These tunnels encapsulate IPv6 traffic within IPv4 packets, and are primarily for communication between isolated IPv6 sites or connection to remote IPv6 networks over an IPv4 backbone. The techniques include using manually-configured tunnels, generic routing encapsulation tunnels, semi-automatic tunnel mechanisms such as tunnel broker services, and fully-automatic tunnel mechanisms such as 6to4 for WANs and intra-site automatic tunnel addressing protocol for the campus environment.
• Deploying IPv6 over dedicated data links. This technique enables IPv6 domains to communicate by using the same Layer 2 infrastructure used for IPv4, but with IPv6 using separate frame relay or ATM permanent virtual circuits, separate optical links, or dense wave division multiplexing.
• Deploying IPv6 over MPLS backbones. This technique allows isolated IPv6 domains to communicate with each other, but over an MPLS IPv4 backbone without modifying the core infrastructure. Multiple techniques are available at different points in the network, but each requires little change to the backbone infrastructure or reconfiguration of the core routers because forwarding is based on labels rather than the IP header itself.

Five steps towards IPv6 adoption by enterprises

• Integrate IPv6 migration with normal product lifecycle replacement. Enterprises can avoid spending money explicitly for IPv6 migration by folding it into planned product procurements from their existing information technology budgets. If IT staffs include upgrades to IPv6 as part of their regular procurement process and select IPv6-capable products from now on, they can take an evolutionary approach to supporting the new protocol.
• Assess existing infrastructure hardware for upgradeability. An enterprise’s top priority should be to take an inventory of its hardware infrastructure that supports only IPv4. Software can be upgraded almost anytime, but some hardware must be capable of supporting IPv6 by design. The devices that are most likely to need attention are routers at the high and low end. High-end routers tend to include acceleration hardware that might be limited to 32-bit addresses. Low-end routers simply may have too little memory to support IPv6 software.
• Use transition technologies judiciously. For routers that must support the IPv6 service by an internal migration deadline but are already slated for replacement shortly thereafter—say, within a few months to one year—enterprises can incorporate tunneling into the network temporarily until the end of the device’s lifecycle. Tunneling involves routing IPv6 packets via virtual paths in the backbone by wrapping them in IPv4 network address headers. Before delivery to end-points, the IPv6 packets are unwrapped and delivered via IPv6 service.
• Integrate IPv6 training into the IT budget and process now. Training might represent a fairly high cost of IPv6 migration, so integrate it into the IT training budget and process as soon as possible. To minimise confusion, IPv6 should be considered a different protocol from IPv4—and one that will take time to learn.
• Add IPv6 to all RFPs. Folding IPv6 support into fresh procurement beyond the core network will help ensure that enterprises meet internal adoption deadlines. This means including IPv6 support in all IT requests for proposals. Doing so avoids the impasse of vendors not making devices IPv6-enabled because RFPs don’t specify it, and of enterprises not upgrading because IPv6 is not available in certain products. The IPv6 deployment process is fairly straightforward. However, employees will need training; routers and operating systems will need updates; management tools will require enhancements; and enterprises will have to deploy IPv6-enabled versions of applications. Integrating IPv6 procurement planning and training into existing IT processes now will help enterprises meet their upgrade deadlines and avoid unexpected and unnecessary costs.