IPV6 MIGRATIION AND ITS CONTINGENCIES, COMPLEXITIES AND IMPLICATIONS

Dr. Hossein Eslambolchi
July 2012

IP Version 6 (IPv6) is the newest version of the Internet Protocol, offering a number of improvements over the existing IP Version 4 (IPv4). Most importantly, IPv6 will provide enough addresses to allow for every region, country, and company to have an abundance of IP Addresses to meet their needs. The IPv4 address space is projected to reach its limits in the 2012-2014 time frame at the current rate of consumption.

IPv6 is expected to coexist with IPv4 for a considerable period of time beyond the limits of IPv4 addresses. The exhaustion of IPv4 address space is a constraint on customer growth and new network and application deployments. It does not directly or immediately impact existing IPv4 installations and customers — unless they need more addresses. With improved and expanded addressing capacity and inherent security and mobility features, IPv6 is a significant improvement over the current protocol. It also facilitates new services based on end-to-end peer-to peer communication models.

Over the past few years, equipment vendors and service providers worldwide have been watching industry trends and testing IPv6 capabilities. Some have programs in place to offer IPv6-based services to enterprises and government agencies. It is clear that service providers must be ready for IPV6 in the 2012-2014 time frame, when we expect to have consumed the last IPv4 address.

 

Reasons to move to IPv6

To date, European Union countries, Japan, Korea, China and India have cited increasing difficulties in obtaining sufficient IPv4 address locations. For historical reasons, the current distribution of IPv4 address allocation is vastly uneven in favor of the United States. These countries believe IPv6 deployment provides a fresh start and serves to distribute IP addresses more equitably. In fact, IPv6 has been adopted as an industry strategy backed by government policies in some of these countries, spurring focused research and development for IPv6 technology and applications. These efforts have, over the past 18 months, made significant in-roads in validating IPv6 multimedia and peer-to-peer (P2P) applications involving new categories of IP-aware devices.

From a U.S. mobile operator perspective, as long as operators continued to assign private IP Addresses to terminals and public addresses to network elements, IPv4 shortage was not thought to be a significant concern for some years to come. However, recent advances in mobile technology and emerging user application trends have highlighted the value of IPv6 for mobile operators and the opportunity costs of not adopting IPv6. The rise of SIP-based real-time IP-Multimedia communications and emerging P2P applications with always-on and always reachable mobile terminal devices have resulted in new thinking about how to leverage IPv6 and ensure the continued growth in data enabled mobile devices.

Visibility of IPv6 in U.S. public policies has also been elevated recently. In response to the U.S. Government’s concern on cyberspace security, the North American IPv6 Task Force (NAV6TF) had promoted end-to-end IPsec based on IPv6 as a means to attainable security solutions. It has since expanded its recommendations calling for IPv6 deployment in the U.S. Government For national business, economic, social and political reasons, starting with the U.S. Department of Defense (DoD).

 

IPv4 Address Space Exhaust

While there are many good reasons to move to IPv6 the driving concern is the imminent Exhaust of IPv4 addresses. When this happens the main impact is the inability to grow the Business. IPv4 has a 32-bit address space providing a theoretical limit of a little over 4 billion Unique addresses. However, past practices of Class (A, B, C) based addressing inadvertently led to address space fragmentation and hugely inefficient allocation and assignment of IP Addresses.

Over the past decade, IP address demands driven by exponential growth of both Wire line Internet, and data-enabled mobile terminals have raised concerns that an IP address Shortage is imminent. Table 1 provides some perspective on the timeframe for exhaustion of IPv4 address inventories at the Regional Internet Registries (RIRs).

 

Government Mandates

While the concern over IPv4 address exhaust will affect all parts of the business in a few years, Government mandates are making service providers move to IPv6 even earlier in those parts of the businesses providing services to government bodies. As noted earlier, countries in the Asia- Pacific region in particular feel as though they have been allocated an inadequate amount of IPv4 address space and are clamoring for a solution. For this reason, the Japanese government mandated the incorporation of IPv6 and set 2005 as the deadline for upgrading existing public and business telephone and data networks. On June 13, 2003, the U.S. Department of Defense Issued a press release mandates any new vendor providing IP services or hardware, for any Project starting after October 2003, to support IPv6. In May 2005,

The US Government Accountability Office issued a report to congress identifying the need to plan for the transition to IPv6. This triggered additional governmental considerations. On August 2nd, 2005, the Office of Management and Budget issued a memorandum calling for all US Federal Government agencies to plan for IPv6 transition with all agencies to be supporting IPv6 on their backbone networks by June 30, 2008.

The National Telecommunications and Information Administration (NTIA) report noted a number of governmental activities in international markets aimed at accelerating deployment of IPv6. Japan, South Korea, China, and the European Commission were identified as having Multimillion dollar projects in place regarding IPv6 deployment.

 

IPv6 Functional Improvements

IPv6 also provides major improvements over IPv4 that can benefit IP mobile and fixed network Providers and network users whether enterprise or consumer. Some of the key new features that come with IPv6 adoption are highlighted below:

 

Improved network management: IPv6 supports Stateless Address Auto-Configuration, which can significantly simplify operator efforts in configuration & management of fixed and mobile Terminals (‘Plug-and-Play’). Site Auto-Renumbering of routers and terminal devices based on Time-scoped public IPv6 addresses facilitates network consolidation.

 

Integrated Security:

Native IPsec support in IPv6 enables robust end-to-end security for Applications. IPsec provides embedded encryption and authentication mechanisms (Encapsulated Security Payload and Authentication Header) for both TCP and UDP.

 

Integrated Quality of Service:

Native Quality of Service (QoS) support via a new Flow Label Field in the IPv6 address header enables robust solutions for delay-sensitive real-time (Conversational) IP-based applications. Further standardization is required and expected to clarify details of this application of the Flow Label.

 

Integrated Mobile IP:

In addition to Layer 2 mobility management by the wireless network, Mobile IP can provide seamless uninterrupted IP sessions via Fast Handovers and Binding Updates between the home address and care-of address as the mobile terminal roams into a foreign network. Mobile IPv6 eliminates the need for the Foreign Agent with auto-configuration and neighbor discovery by the mobile host in the foreign network. Direct routing of the Forwarded traffic to the mobile host, (i.e. avoidance of triangular or ‘trombone’ routing) is also Supported. Mobile IP is required for seamless handovers when users move between access Technologies, e.g. 3GPP to WLAN. There are a number of features to mobility, but some form of Mobility is inherent in the Next Generation Network concept of accessing any application from anywhere.

 

Improved Routing:

Route aggregation capability similar to the classless inter-domain routing (CIDR) in IPv4 is incorporated in IPv6 routing. The Uncast and Multicast routing of IPv4 have been extended with any cast routing capabilities.

 

Ad-Hoc Networking:

IPv6 will support the end-to-end (E2E) routing and addressing Requirements of emerging Ad Hoc networks, whereby mobile wireless devices can establish Communications anytime and anywhere without the aid of a central infrastructure (i.e. base Station or access point). Mobile devices can act as both a router and host either in mobile ad-hoc networks (MANET) or personal area networks (PAN) that can inter-work with the cellular Network or the Internet. Such applications, made possible by short-range radio technologies (E.g. Bluetooth and 802.11) could extend the reach of conventional cellular networks both in the home networking and in remote out-of-reach scenarios (e.g. sensors located in remote areas for Environmental monitoring, where information can be relayed by roaming mobile nodes acting as Multi-hop routers).

 

1.5 Leveraging IPv6 for New Applications

The main feature of IPv6 is the virtually limitless address space with 340,282,366,920,938,463,463,374,607,431,768,211,456 addresses compared to ~4 billion with IPv4. This can provide persistent public IP addresses to a virtually unlimited number of always-on devices ranging from currently well-known device categories such as PCs and mobile Phones to emerging device types such as sensor networks and intelligent infrastructures. The Integrated support of IPsec deploys capabilities that can be leveraged for improved security.

 

This large address space enables a variety of new network capabilities and services:

Push applications (e.g., push emails/messaging and alerting services)

Peer-to-Peer based applications

Seamless support of IP mobility

 

IPV4 to IPV6 Areas of Consideration

The transition to IPv6 will take several years meaning that IPv4 and IPv6 will coexist for a long Time. To aid in the transition and the coexistence, IETF has defined several transition mechanisms: dual stack, tunneling, and translation. The dual stack mechanism enables a Device to support both IPv4 and IPv6 protocol stacks simultaneously. Dual stack is also the Base mechanism that enables tunneling.

The tunneling mechanism enables one IP version to be encapsulated in and transported in the packets of another IP version. The third transition mechanism is translation which translates one protocol into the other protocol. The architecture guidelines provide directions on the applicability and use of these transition mechanisms in different scenarios and also describe key security considerations. The dual stack mechanism will be the most widely used transition tool in Service Provider’s networks, service infrastructures, and customer devices.

Tunneling will be used where needed to enable IPv6 to be encapsulated in IPv4 and sent over an IPv4 network. Additional guidance is provided as to which types of tunnels are allowed (configured and negotiated tunnels) and which types should be avoided (automatic tunnels). Translation is the least desirable mechanism and should be avoided or used in a limited manner when there are no other alternatives. Translation breaks the end-to-end model which impacts security and QoS. Service Providers will likely have some translation in the network to support IPv4 to IPv6 Inter-working.

An IPv6 Addressing Strategy and Plan needs to be developed. This provides high level Addressing policies and address assignment plan for some key IP services such as Managed Internet Services (MIS), Virtual Private Networks (VPNs), FTTH high speed access, DSL, and Mobility. The addressing policies are based on IETF and Regional Internet Registries (RIRs) recommendations.

The assignment plan for each service often has what is referred to as a Wide Area Network (WAN) component and a Local Area Network (LAN) component. The WAN interface connects the customer to the Service Provider’s network and therefore the address assignment is preferred to be not globally routable for security reasons.

The LAN address assignment is for customers to connect to Internet or to other network and therefore is generally preferred to be globally routable IP addresses. Further study is needed to make a final determination on addressing policies.

In developing the IPv6 transition roadmap, the key IP infrastructures needs to grouped into the Following categories: IP/MPLS Transport, Access/Aggregation Network, Service infrastructure, Customer Devices and Network Services.

Within each infrastructure category, there are a number of functional areas. For each functional area, an IPv6 impact assessment and an initial IPv6 transition roadmap are provided. The overall transition strategy is to enable dual stack IPv4/IPv6 on network and service infrastructures so that we can service IPv6-only customers while continuing to serve IPv4 customers during the transition period to IPv6. For the customer devices, we plan to start supporting IPv6-only service so that we can stop assigning IPv4 addresses to customers when such IPv4 addresses are exhausted. Since Network and Service Infrastructures will be dual stack (IPv4/IPv6 capable) for a long time during the transition period,

It is assumed that network management can stay on IPv4 for a while other than changes that are needed to support IPv6-only customers are implemented.

IP/MPLS Transport provides the underlying IP connectivity for customers to reach the Internet, Their own IP networks and Service Provider’s IP services. The IP/MPLS Transport category covers the IP/MPLS network and the service networks such as Edge Router, Network Aggregation Router, VoIP Network, and Hosting Router that support various Service infrastructures.

 

The overall transition strategy for the IP/MPLS Transport Infrastructure is to transition the IP/MPLS Network to support IPv6 and then extend the transition out to the Service Networks. The IP/MPLS Core will remain on IPv4 for a long time because the Core switches packets at the MPLS layer and not the IP layer; furthermore, we do not expect we will run out of IPv4 addresses for infrastructure usage. However, the IP/MPLS Core does need to recognize some parameters in the IPv6 header to support network monitoring and security functions. The IP/MPLS Edge and the Service Networks will be made IPv4/IPv6 dual stack capable to support IPv6-only customers as well as IPv4 customers.

The Access network infrastructure provides customers with different access mechanisms to reach the IP/MPLS backbone network. This category covers a wide range of access networks, from wire line networks, DSL to wireless networks such as Wi-Fi, WiMAX and LTE Mobility.

A few have minimal to no impacts due to IPv6 because they do not interface with the Customers via an IP interface such as Metro Ethernet, Mobility Cell Site Transport, and Mobility Radio Access Network. Therefore, the transition roadmap and the urgency to upgrade to IPv6, Varies among the different access networks. In general, for those access networks that are impacted by IPv6, it is recommended to use dual stack or tunnel mechanisms to gain initial IPv6

Experience and then eventually evolve the infrastructure to support IPv6-only customers while continuing to support IPv4 customers. Following is the assessment for each of the access Network:

The FTTH/GPON IP Network will first need to implement dual stack High Speed Internet Access (HSIA) service to gain IPv6 experience. Then, Service Providers will need to implement IPv6-only for high speed access and VoIP services with the possibility of IPTV service staying on IPv4 using private IPV4 address in case the IPTV service platform is not able to support IPv6 Until after the exhaust date. Eventually, the plan is to move IPTV service to IPv6 as well. This clearly depends on vendor support for IPV6 for GPON/FTTH technologies.

DSL upgrade to IPv6 will depend on IP-DSLAM footprint and DSL Subscriber forecast. We will first use host based tunneling for initial customer trials to gain IPv6 experience. Then we will implement dual stack service and finally move to supporting IPv6-only customers.

Remote Access relies on the underlying access networks (DSL, WiFi, Mobility, 3rd Party network, etc.) That connects customers to the target network. Therefore, we will first implement dual stack on the Remote Access Client and Access Gateways to Tunnel IPv6 over IPv4. As the underlying access networks transition to IPv6, Remote Access will then be able to support IPv6-only Remote Access service. The Associated elements (e.g. AAA) and support systems do not need to Support IPv6 forwarding but do need to recognize some of the IPv6 related Parameters. As we move to IPv6-only Remote Access service, some of the associated elements and systems will become IPv6 capable.

WiFi network assigns both private and public IPv4 addresses to customers Therefore IPv4 address exhaustion will not have an immediate impact on customers because we can leverage private IPv4 addresses. However, as we move toward an IPv6 dominant network or if we want to offer new service features (e.g. peer-to-peer, always-on applications) that leverage IPv6 capabilities, we will need to move WiFi to IPv6. A possible scenario is to use IPv6 over IPv4 tunnel initially, then dual stack, and finally IPv6-only. The WiFi network relies on the DSL network for IP connectivity so we need to ensure that the underlying DSL network is also IPv6 ready.

The Metro Ethernet network provides Layer 2 services and Ethernet Access to Layer 3 services. We do not expect major IPv6 impacts here. However, if the Metro Ethernet network needs to mark traffic based on customer’s IP header, then The Metro Ethernet network elements will need to be IPv6 aware to differentiate the IPv4 header from the IPv6 header to examine the proper IP QoS bits.

Mobility Cell Site Backhaul has no IPv6 impacts. It uses private IPv4 addresses and these can be reused locally within the Radio Access Network (RAN). However, we do need to understand better if Femto, Long Term Evolution (LTE), and changes In OA&M network will have impacts on the mobility cell site backhaul.

Mobility Radio Access Network has no IPv6 impacts. It uses private IPv4 Addresses and these can be reused locally within the RAN. Similar to Mobility Cell Site Backhaul, we need to understand better if Femto, LTE, and changes in the OA&M network will have impacts on the Mobility Radio Access Network.

 

· Mobility Packet Core supports Mobility data services and has major IPv6 impacts. But, we do not need to upgrade the whole infrastructure right away. Initially, we will need to support IPv6 user traffic while keeping the packet core transport on IPv4 and Upgrade only the Gateway GPRS Support Node (GGSN) to dual stack to interface with external IPv6 networks. Service Providers will need to use a strategy to do this. One option is to enable IPv6 one service at a Time. As more and more services move onto IPv6, we will then evolve the whole Mobility packet core infrastructure to IPv6.

Service Infrastructures consist of servers and applications that provide customers with Service functionalities. This category covers VoIP, Internet Hosting, IPTV, and Security Services, Content Distribution, and Mobility Services. The overall strategy here should be to upgrade the service infrastructure to dual stack (IPv4/IPv6) to support IPv6-only customers and continue to support IPv4 customers. New service infrastructures must support IPv6 dual stack from the very beginning. Growth services will migrate to IPv6 one by one and will drive the underlying infrastructures (network, service, systems, etc.) to migrate to IPv6.

 

Customer Devices fall into three categories: consumer wire line devices, managed devices for Enterprises and mobility devices. The customer device will play a critical role in ensuring a Good customer experience Service Providers transitions to IPv6. This is also one of the most important areas to address because consumer devices are the largest consumer of public IPv4 Addresses. The overall IPv6 transition strategy for customer devices is to begin deploying Customer devices that can support both IPv4 and IPv6 so that depending on the network and Service infrastructure that the customer device is connected to, the appropriate IP protocol stack can be enabled.

 

Finally, Network Services provides customers with a set of protocols and applications for Authentication, IP address assignment, name resolution, and consistent network based timing for end user applications. This category covers AAA, DNS, DHCP and NTP.

AAA is the mechanism for Authentication, Authorization and Accounting for Subscribers on the various networks. Authentication is the user login and password information. Authorization provides the information on what the user is allowed to do on the network or Resources the user is allowed to access. Accounting is the mechanism for billing Data, logins, auditing information and other traffic or access records. The primary IPv6 impact on AAA servers is the need to upgrade the RADIUS application on several of the networks to support IPv6.

DNS or Domain Name Services is the method to resolve network resource names Into IP addresses. The DNS infrastructure will see some impact from IPv6 But the systems in place today is already mostly capable of supporting IPv6.

DHCP or Dynamic Host Configuration Protocol is a system for dynamically assigning IP addresses to subscribers on a network. Currently DHCP is most widely used on The FTTH/GPON network to assign IP addresses. The current DHCP systems in almost all service providers are not capable of supporting IPV6 so this will need to be upgraded.

 

IPV6 KEY POINTS:

• Internet Protocol Version 4 (IPv4) has been a standard since 1981, and currently forms the foundation of most Internet transactions.

 

• IPv4 suffers from several important shortfalls, chief of which include a lack of sufficient address space (fewer than four billion addresses, a problem compounded by inefficient allocation), as well as inadequate security, multicasting, and Quality of Service (QoS) features.

• A next generation Internet Protocol Version 6 (IPv6), sometimes called IPng, now exists and offers a number of improvements:

  • Virtually unlimited address space
  • Improved security
  • Improved multicasting/any casting features
  • Improved QoS features

• Methods devised to overcome IPv4 shortcomings, such as Network Address Translation (NAT), often introduce obstacles to seamless Internet connectivity.

• The cost and complexity of upgrading from IPv4 to IPv6 is significant. Methods of coexistence (e.g., thru tunneling) exist. A gradual transition is anticipated.

• The greatest impetus for adoption of IPv6 comes from the Asia Pacific region, owing to the acute shortage of allocated IPv4 addresses, as well as U.S. Department of Defense (DOD).

• Internet Protocol Version 4 (IPv4) has been used to run the Internet for more than 20 years. Demand for Internet addresses and interest in advanced applications that can benefit from additional features in the protocol are making the IPv4 address space, which theoretically supports fewer than four billion Internet addresses — far too few to meet the need.

• Methods to prolong the use of IPv4 have been devised, such as Network Address Translation (NAT), allowing multiple endpoints to share one IPv4 address. While this industry supported “workarounds” have extended the availability of addresses, there are other limitations created in using NAT for some important applications, such as peer-to-peer applications and potentially other future applications.

• Work began in the mid-1990s to develop a “next generation” Internet protocol. The result, IPv6 (sometimes called IPng), was approved as a standard in 20001. IPv6 brings several benefits: it supports a vast number of Internet addresses, enough to meet all current needs; in addition, a fully implemented IPv6 offers the ability for applications providers to offer users with improved performance, guaranteed quality of service capabilities, manageability, scalability, data protection, and multicast support.

• Regions outside of North America are already feeling the shortage of IPv4 addresses. Countries that were early adopters of the Internet have established a base of IPv4 addresses; however, countries which were late to adopt the Internet and countries with vast potential

IPv6 General Model:

Populations of Internet and mobile web users are faced with a lack of sufficient Internet addresses.

• The shortage of IPv4 addresses in Asia is driving the introduction of IPv6 technology there. Japan, South Korea, and China have federal mandates and incentives for the private sector to adopt IPv6 on an accelerated schedule. The European Union has mandated that, in the near future, network devices must support IPv6.

• The cost and complexity of upgrading from IPv4 to IPv6 is significant. More development is required to smooth the transition and coexistence between legacy IPv4 networks and IPv6 networks to provide businesses and suppliers a smooth transition path. These issues, and the workarounds extending the life of IPv4, have delayed IPv6 implementation, especially in the United States.

• Governmental adoption can play a role in speeding adoption of IPv6. The greatest impetus for the movement by governmental agencies to IPv6 in the U.S. comes from the U.S. Department of Defense. A further announcement of a commitment strategy to support migration to IPv6 has been presented by the U.S. Department of Commerce.

• The world’s largest IPv6 network was set up in March 2004 in the Moonv6 Project. Carriers, Sprint and NTT were among the ISPs supporting this network whose goal is to provide a platform for testing, training and software development for IPv6.

• Some analysts believe that IPv4 and IPv6 should coexist as mandatory standards until 2007, with IPv4 not totally replaced until some years following. However, Carriers’ view is that many

Customers will be slow to move to IPv6; in fact, it is quite possible that IPv4 will never be fully replaced.2 it is quite likely that IPv6 will need to operate in parallel with IPv4, perhaps indefinitely. Accordingly, the need to both educate customers about the benefits of moving to IPv6 and support legacy IPv4 applications should be factored into Carriers’ migration strategy and marketing plans, and into both domestic and global public policy initiatives, domestically and globally.

• IPv6 is beginning to receive significant attention as a key policy issue within the Internet community. Accordingly, needs to develop a public policy position with a public policy statement that encompasses a defined a migration strategy. This statement should establish OPERATORS technological commitment and should assure customers and public policy makers of OPERATORS’s leadership in this area of advancing IP networking.

3.2 THE TECHNOLOGY:

For more than 20 years the Internet has been based on Version 4 of the Internet Protocol (IPv4). IPv4 was designed to accommodate approximately four billion potential Internet addresses using 32-bits each, which seemed more than adequate back in the 1980s.3 As the Internet, grew, blocks of Internet addresses were assigned to various organizations and countries. However, by 1995 and 2000 about one-quarter and one-half of all potential Internet addresses were taken,

 

Not all bit strings of length 32 are valid Internet addresses. IPv4 has three classes of addresses, Class A, Class B, and Class C, and also supports services such as multicasting. Because of the way these addresses are specified, only around 3 billion IPv4 addresses are available for use by Internet endpoints. Moreover, because the way IPv4 addresses are assigned to endpoints, in practice, only about 250 million IPv4 addresses can be used respectively.

 

IPv6 Growth in 2020 Decade:

With the fast growth in the number of Internet devices, it looked as if IPv4 addresses would be exhausted. However, researchers and technologists in the Internet standards organizations anticipated this problem in the early 1990s and developed methods to extend the life of IPv4. Meanwhile, they also initiated the development of the “next generation” of IP, an effort that led to the development and standardization of IPv6.

The principal methods developed for prolonging the use of IPv4 include network address translation (NAT), classless inter-domain routing (CIDR), and PPP/DHCP address sharing. Using NAT, a single device, such as a router, acts as an agent between the Internet and a local network, so that only a single IP address is required to represent an entire group of endpoints. Although the main motivation of CIDR was to reduce the size of routing tables carried by ISPs, it also permits smaller allocations of addresses to customers and ISP. In particular, it lets a grouping of separate IP networks appear as part of a single subnet, allowing service providers to conserve addresses by divvying up pieces of a full range of IP addresses to multiple customers.

Besides these technical approaches for further extending the life of IPv4, unused, but assigned, addresses have been reclaimed for use. Another major driver is action by the Regional Internet Registries to prevent waste of IP addresses. PPP and DHCP are more auto configuration aids; as more and more people move to always-on broadband connections, the address-conservation aspects of this will decrease.

Unfortunately, the conservation of IPv4 address has undesirable effects that penalize performance and increase operating costs. The measures used make system administration more complex and error prone. In particular, to configure NAT to support remote administration entails high operating costs. The lack of transparency of NAT makes reliable diagnoses of problems difficult. When NAT is used, the on-the-fly manipulation of IP packet headers, necessary for establishing a link between a private network and the public network, makes end-to-end IPSec security impossible, as NAT’s modification of packet headers leads to a rejection of packets during IPsec controls.

Moreover, NAT degrades performance, which is especially important for applications sensitive to transit times. Perhaps worst of all, NAT is a stumbling block for launching peer-to-peer applications, which have recently emerged as key applications for both end users and businesses. For such applications, it is necessary to know the correspondent address in the private network, requiring complex application-related mechanisms for locating the address of the final correspondent. Finally, installing a local web server causes problems because the server must be accessed from the outside via NAT.

While the interim measures taken to prolong the life of IPv4 has created other problems, the most immediate challenge driving migration to a new protocol is that the number of Internet endpoints is growing explosively, with the result that it is doubtful that IPv4 can meet future needs.

Technologists in many industry sectors are predicting or actually designing complex applications for devices that will require IP addresses for a vast number of endpoints.

More than one billion PCs, more than one billion mobile Internet endpoints (including mobile phones), more than one billion cars, billions of home-based voice-over-IP gateways, as well as the growing numbers of gaming stations and home appliances that may each need their own Internet addresses. By the early 1990s researchers anticipated the need for more Internet addresses and began work on a new generation of the Internet protocol. The result is Internet Protocol Version 6, IPv6.

IPv6 supports 128-bit addresses and the number of addresses available for Internet endpoints is vast4, exceeding the number needed for any scenario yet devised. IPv6 was approved by the IETF as a Proposed Standard in 1995 and was approved as a standard in 2000.

Although the original impetus for the creation of IPv6 was to increase the address space, the opportunity to design a new Internet Protocol made it possible to introduce additional enhancements. IPv6 was designed with architecture more consistent than that of IPv4, with 4 although not all 2128 bit strings of length 128 are valid Internet addresses, the number of available IPv6 addresses is vast.

 

IPv6 Improved Security:

Hooks that can be used for improved support for improved security, multicasting and any casting, and mobility, as well as potentially for quality of service. One particular security-related advantage of IPv6 is that worms to an extent will be harder to write for IPv6 networks. Another is the inclusion of the secure neighbor discovery protocol, which protects neighbor discovery messages through the use of cryptographically generated addresses.

 

Although IPv6 brings many benefits, it is not a panacea for all challenges the Internet faces. IPv6 does add several layers of built-in security and once employed, it will help to stop certain classes of attacks by making it difficult to spoof, or masquerade, as a different computer. However, IPv6 has no ability to close most known network vulnerabilities, which usually exploit security weaknesses above the IP layer. Yet it is a key part of the solution to establishing a “culture of networking security”.

 

IPv6 enables better traffic flow than IPv4 and enables automatic connectivity. In particular, IPv6 offers “neighbor” discovery and address auto configuration capabilities supporting mobility, allowing hosts to operate anywhere without special support. Using these capabilities, a host can be reached no matter where it is connected to the Internet. This is accomplished by binding the current “care-of address” of a mobile host to its home address.

 

Although IPv6 offers advantages, its adoption has been slowed by the complexity and costs associated with the move from IPv4 to IPv6. Because of this, the migration from IPv4 to IPv6 will require the coexistence of these protocols for some time. The key strategies used in deploying IPv6 at the edge of such a network involve carrying IPv6 traffic over the IPv4 network, allowing isolated IPv6 domains to communicate with each other before the full 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 protocol. All techniques allow networks to be upgraded, and IPv6 to be deployed incrementally with little to no disruption of IPv4 services.

 

The key strategies are to deploy IPv6 over IPv4 tunnels that encapsulate IPv6 traffic within IPv4 packets, to deploy IPv6 over dedicated data links, to deploy IPv6 over MPLS IPv4 backbones, and to deploy IPv6 using dual-stack backbones allowing IPv4 and IPv6 applications to coexist in a dual IP layer routing backbone. Some of the Japanese ISPs have found it better to run parallel networks, especially given the relatively immature support for v6 in routers.

 

Tier 1 Carriers Market Interest in IPv6:

Network equipment vendors offer a comprehensive set of offerings, including routers, switches, and firewalls that support IPv6. Computer hardware and software vendors are aggressively pursuing IPv6. Every significant computer operating system and almost every network equipment vendor now has IPv6 support. Although hardware and software vendors have enabled the adoption of IPv6, the migration to IPv6 will take some time, with the migration occurring at different rates in different parts of the world.

 

The heaviest demand for new IPv6 addresses is in Asia, because that continent has struggled with a minimal allocation of IPv4 addresses. (For example, U.S.-based Level 3 Communications has about 48 million unique IP addresses from the allocation of three Class A domains, which is almost as many IPv4 addresses as have been allocated to the entire continent of Asia.)

 

Enterprises in Asia are planning to adopt IPv6 technology because no IPv4 addresses are available to meet their current and future needs. Moreover, Japan, South Korea, and China have federal mandates and incentives for the private sector to adopt IPv6 on an accelerated schedule. IPv6 nodes use the Neighbor Discovery Protocol (NDP) to discover other nodes on the link, to determine their link-layer addresses to find routers, and to maintain reachability information about the paths to active neighbors. If not secured, NDP is vulnerable to various attacks. See: RFC 3971.

 

IPv6 Migrations Considerations:

China is testing IPv6 networks in some big cities around the country. Japan has already implemented an IPv6 production network, which is used by every service provider in the country.

 

South Korea is working with the European Union to develop applications and services using IPv6. Not only does IPv6 provide a way that the shortage of Internet addresses in Asia can be met, but there is a growing opinion that early adoption of IPv6 offers a competitive advantage for Asia relative to the U.S. for technological leadership in the Internet. This view is also held within the European Union.

 

The migration to IPv6 is slower in Europe than in Asia, but the European Union has mandated that in the next few years, network devices must support IPv6. Also, in January 2004, the EU launched a large research IPv6 network. Currently, IPv6 is used extensively on several large research networks in Europe and Asia. Commercial IPv6 service is available in Japan, Korea, Malaysia, Taiwan, Hong Kong, and Australia, as well as in Europe, including the U.K., the Netherlands, France, Germany, and Spain.

 

There has also been a lot of interest in IPv6 in the U.S. among the research and vendor communities, but the high cost of rolling it out has deterred service providers from introducing it. Implementing IPv6 requires replacing the IP stacks on routers, switches, and other networking equipment and supporting IPv6 on servers, hosts, and other end devices. Some industry analysts have said that it will be a long time before IPv6 will be used by North American carriers, as measures such as NAT extend the life of IPv4.

 

The greatest impetus for the movement to IPv6 in the U.S. comes from the internet mobility and U.S. Department of defense and government agencies worldwide.

 

A key area for the adoption of IPv6 is for use by digital mobile devices. The use of IPv6 is mandated in Release 8 of the LTE Generation Partnership Project (3GPP), which develops standards for advanced mobile networks. Specifically, UMTS Release 5 mandates IPv6 in all handsets and the 3G Internet Multimedia Subsystem is defined to run only on IPv6.

 

 

THE PLAYERS:

Consortia and research projects are playing a major role in driving the adoption of IPv6. Among the most important of these are:

 

Internet2, a consortium led by over 200 universities working in partnership with industry and government to develop and deploy advanced network applications and technologies, is actively deploying IPv6 on campus, regional and backbone networks.

 

POTENTIAL IMPACTS:

Supporting IPv6 should be an important part of future strategy for providing IP services.

 

In particular, companies should have a plan for IPv6 that can be executed as demand arises. It is also important for to demonstrate activity in IPv6 trials to be competitive with major competitors.

 

OPERATORS should also have a public policy position and transition plan that can be provided to public policy makers, such as the Department of Commerce, DoD, global policy forums important to OPERATORS ’s global IP networking business, large customers and business suppliers. OPERATORS should consider providing IPv6 services in the U.S. and possibly throughout Asia and Europe as well. Many recent bids from large corporations have asked for Carriers’ IPv6 plan.

 

Several large customers, such as IBM and Apple, have strong pushes for IPv6. Furthermore, it will become crucial for to offer IPv6 services for U.S. government contracts, especially for the U.S. Department of Defense, especially because the U.S. DoD has mandated support for IPv6 and is planning a transition to it.

 

OPERATORS have done a limited amount of work on IPv6. It has procured a block of IPv6 addresses and has developed a draft policy for address assignment. It also has done IPv6 prototyping and testing and has participated in the Moon6 Project. There are also proposals for developing a high-level IPv6 service plan.

 

This should include plans for OSSs to support IPv6 services. It would also be worthwhile for to work to create an “IPv6 community” involving existing customers and business partners. This will provide customers a reason to begin using IPv6. It will important for to continue to strongly support IPv4 because the Internet today runs on IPv4 and will continue to run on IPv4 for the foreseeable future. In addition, it also needs to provide tools that help our customers use both IPv4 and IPv6, and to begin using IPv6 in their enterprise, such as on a VPN extranet.

 

OPERATORS could partner with equipment vendors to help its customers make this migration and could educate its customers as to the benefits of IPv6, beyond just the availability of additional addresses. By aggressively supporting IPv6, might be able to win new business from enterprises currently using who might be slower to move to IPv6, and will certainly be more competitive within the government marketplace.

 

OPERATORS also need to work closely with equipment vendors to get full IPv6 support. Many vendors OPERATORS uses still do not support IPv6, and even ones that do (e.g., Cisco) only support it with various limitations on performance and features. Currently, the only major router vendor to provide full support for IPv6 is Juniper Networks. Supporting IPv6 will have major systems Potential IMPACTS, as so many of the systems supporting IP service rely on a 32-bit address as a key into a database. Many of the rules used by the industry, including Carriers, to manage IPv4 addresses will also need to change in an IPv6 environment. Consequently, an IPv6 service plan needs to be mapped out that will identify the range of services may want to run over IPv6, including inter-working services with IPv4 and new services such as any casting. Network development plans should be formulated as well as implementation plans needs to work actively with the relevant standards groups to ensure the evolution of standards in this area that are adopted broadly across the global network provider/supplier base.