Advocacy for IPv6 Deployment—Example

0
194

We include below some excerpt from the European Economic and Social Committee and the Committee of the Regions [39] to emphasize the issues related to IPv6. Clearly, issues about IPv6 impact not only Europe but the entire world.

The European Economic and Social Committee and the Committee of the Regions has issued an “Action Plan for the deployment of IPv6 in Europe.” It is the objective of this Action Plan to support the widespread introduction of the next version of the IP (IPv6) for the following reasons:

  • Timely implementation of IPv6 is required as the pool of IP addresses provided by the current protocol version 4 is being depleted.
  • IPv6 with its huge address space provides a platform for innovation in IP based services and applications.

Preparing for the Growth in Internet Usage and for Future Innovation. One common element of the Internet architecture is the IP that in
essence gives any device or good connecting to the Internet a number, an address, so that it can communicate with other devices and/or goods. This address should generally be unique, to ensure global connectivity. The current version, IPv4, already provides for more than 4 billion such addresses. Even this, however, will not be enough to keep pace with the continuing growth of the Internet. Being aware of this long-term problem the Internet community developed an upgraded protocol, IPv6, which has been gradually deployed since the late 1990s.

In a previous Communication on IPv6, the European Commission made the case for the early adoption of this protocol in Europe. This Communication has been successful in establishing IPv6 Task Forces, enabling IPv6 on research networks, supporting standards, and setting-up training actions. Following the Communication, more than 30 European R&D projects related to IPv6 were financed. Europe has now a large pool of experts with experience in IPv6 deployment. Yet, despite the progress made, adoption of the new protocol has remained slow while the issue of future IP address scarcity is becoming more urgent.

Increasing Scarcity of IPv4 Addresses: A Difficulty for Users, an Obstacle to Innovation. Initially all Internet addresses are effectively held
by the IANA and then large blocks of addresses are allocated to the five RIRs that in turn allocate them in smaller blocks to those who need them, including ISPs. The allocation, from IANA to RIR to ISP, is carried out on the basis of demonstrated need: there is no preallocation.

The address space of IPv4 has been used up to a considerable extent. At the end of January 2008 about 16% was left in the IANA pool, that is, approximately 700 million IPv4 addresses. There are widely quoted and regularly updated estimates that forecast the exhaustion of the unallocated IANA pool somewhere between 2010 and 2011. New end users will still be able to get addresses from their ISP for some time after these dates, but with increasing difficulty.

Even when IPv4 addresses can no longer be allocated by IANA or the RIRs, the Internet will not stop working: the addresses already assigned can and most probably will be used for a significant time to come. Yet the growth and also the capacity for innovation in IP-based networks would be hindered without an appropriate solution. How to deal with this transition is currently the subject of discussion in the Internet community in general, and within and amongst the RIR communities in particular.

All RIRs have recently issued public statements and have urged the adoption of IPv6.

IPv4 is only a Short-Term Solution Leading to More Complexity. Concerns about the future scarcity of IP addresses are not a recent phenomenon. In the early days of the Internet, before the establishment of the RIRs and before the take-off of the World Wide Web, addresses were assigned rather generously. There was a danger of running out of addresses very quickly. Therefore, changes in allocation policy and in technology were introduced that allowed allocation to be more aligned to actual need.

One key IPv4 technology has been NAT. NATs connect a private (home or corporate) network that uses private addresses to the public Internet where public IP addresses are required. Private addresses come from a particular part of the address space reserved for that purpose. The NAT device acts as a form of gateway between the private network and the public Internet by translating the private addresses into public addresses. This method therefore reduces consumption of IPv4 addresses. However, the usage of NATs has two main drawbacks, namely:

  • It hinders direct device-to-device communication: intermediate systems are required to allow devices or goods with private addresses to communicate across the public Internet.
  • It adds a layer of complexity in that there are effectively two distinct classes of computers: those with a public address and those with a private address. This often increases costs for the design and maintenance of networks, as well as for the development of applications.

Some other measures could extend the availability of IPv4 addresses. A market to trade IPv4 addresses might emerge that would offer incentives to organizations to sell addresses they are not using. However IP addresses are not strictly property. They need to be globally acceptable to be globally routable, which a seller cannot always guarantee. In addition, they could become a highly priced resource. So far, RIRs have been skeptical about the emergence of such a secondary market. Another option consists of trying to actively reclaim those already-allocated address blocks that are underutilized. However, there is no apparent mechanism for enforcing the return of such addresses. The possible cost of it has to be balanced against the additional lifetime this would bring to the IANA pool. Though such measures may provide some interim respite, sooner or later the demand for IP addresses will be too large to be satisfied by the global IPv4 space. Efforts to stay with IPv4 too long risk increasing unnecessary complexity and fragmentation of the global Internet. A timely introduction of IPv6 is thus the better strategy.

IPv6: The Best Way Forward. IPv6 provides a straightforward and long-term solution to the address space problem. The number of addresses defined by the IPv6 protocol is huge. IPv6 allows every citizen, every network operator (including those moving to all IP “Next Generation Networks”), and every organization in the world to have as many IP addresses as they need to connect every conceivable device or good directly to the global Internet. IPv6 was also designed to facilitate features that were felt to be missing in IPv4. Those features included quality of service, autoconfiguration, security, and mobility. In the meantime, however, most of those features have been engineered in and around the original IPv4 protocol. It is the large address space that makes IPv6 attractive for future applications as this will simplify their design when compared to IPv4. The benefits of IPv6 are, therefore, most obviously apparent whenever a large number of devices or goods need to be easily networked, and made potentially visible and directly reachable over the Internet. A study funded by the Commission demonstrated this potential for a number of market sectors such as home networks, building management, mobile communication, defense and security sector, and car industry.

Prompt and efficient adoption of IPv6 offers Europe potential for innovation and leadership in advancing the Internet. Other regions, in particular the Asian region, have already taken a strong interest in IPv6. For instance, the Japanese consumer electronics industry increasingly develops IP enabled products and exclusively for IPv6. The European industry should therefore be ready to meet future demand for IPv6-based services, applications, and devices and so secure a competitive advantage in world markets.

To conclude, the key advantage of IPv6 over IPv4 is the huge, more easily managed address space. This solves the future problem of address availability now and for a long time to come. It provides a basis for innovation—developing and deploying services and applications that may be too complicated or too costly in an IPv4 environment. It also empowers users, allowing them to have their own
network connected to the Internet.

What Needs to be Done? IPv6 is not directly interoperable with IPv4. IPv6 and IPv4 devices can only communicate with each other using
application-specific gateways. They do not provide a general future-proof solution for transparent interoperability. However, IPv6 can be enabled in parallel with IPv4 on the same device and on the same physical network. There will be a transition phase (expected to last for 10, 20, or even more years) when IPv4 and IPv6 will coexist on the same machines (technically often referred to as “dual stack”) and be transmitted over the same network links. In addition, other standards and technologies (technically referred to as “tunneling”) allow IPv6 packets to be transmitted using IPv4 addressing and routing mechanisms and ultimately vice versa. This provides the technical basis for the step-by-step introduction of IPv6. Because of the universal character of the IP, deployment of IPv6 requires the attention of many actors worldwide. The relevant stakeholders in this process are as follows:

  • Internet organizations (such as ICANN, RIRs, and IETF) that need to manage common IPv6 resources and services (allocate IPv6 addresses, operate DNS servers, etc.), and continue to develop needed standards and specifications. As of May 2008, the regional distribution of allocated IPv6 addresses is concentrated on Europe (R´eseaux Internet Protocol Europ´eens or RIPE: 49%), with Asia and North America growing fast (Asia–Pacific Network Information Centre, APNIC: 24%; ARIN: 20%). Less than half of those
    addresses are currently being announced on the public Internet (i.e., visible in the default-free routing table). In the DNS the root and top-level name servers are increasingly becoming IPv6 enabled. For instance, the gradual introduction of IPv6 connectivity to. eu name servers started in 2008.
  • ISPs that need over time to offer IPv6 connectivity and IPv6 based services to customers: There is evidence that less than half of the ISPs offer some kind of IPv6 interconnectivity. Only a few ISPs have a standard offer for IPv6 customer access service (mainly for business users) and provide IPv6 addresses. The percentage of “Autonomous Systems” (typically ISPs and large end users) that operate IPv6 is estimated at 2.5%. Accordingly, IPv6 traffic seems to be relatively low. Typically the IPv6/v4 ratio is less than
    0.1% at Internet Exchange Points (of which about one in five supports IPv6). However, this omits direct ISP to ISP traffic and IPv6 that is “tunneled” and so appears at first glance to be still IPv4. Recent measurements suggest that this kind of traffic IPv6 that is “tunneled” is growing.
  • Infrastructure vendors (such as network equipment, operating systems, network application software) that need to integrate IPv6 capability into their products: Many equipment and software vendors have upgraded their products to include IPv6. However, there are still issues with certain functions and performance, and vendor support equivalent to IPv4. The installed equipment base of consumers, such as small routers and home modems to access the Internet, still by and large do not yet support IPv6.
  • Content and service providers (such as websites, instant messaging, email, file sharing, voice over IP) that need to be reachable by enabling IPv6 on their servers: Worldwide there are only very few IPv6 websites. Almost none of the global top sites offer an IPv6 version. The de facto nonexistence of IPv6 reachable content and services on the Internet is a major obstacle in the take-up of the new protocol.
  • Business and consumer application vendors (such as business software, smart cards, peer-to-peer software, transport systems, sensor networks) that need to ensure that their solutions are IPv6 compatible and increasingly need to develop products and offer services that take advantage of IPv6 features. Today, there are few, if any, current applications that are exclusively built on IPv6. One expectation has been that proliferation of IP as the dominant network protocol would drive IPv6 into new areas such as logistics and traffic management, mobile communication, and environment monitoring that has not taken place to any significant degree yet.
  • End users (consumers, companies, academia, and public administrations) that need to purchase IPv6 capable products and services and to enable IPv6 on their own networks or home Internet access: Many home end users, without being aware of it, operate IPv6 capable equipment and yet, as a result of missing applications, without necessarily making use of it. Companies and public administrations are cautious to make changes to a functioning network without a clear need. Therefore not much user deployment in private networks is visible. Among the early adopters have been universities and research institutions. All EU national research and education networks also operate on IPv6. The European G´eant network is IPv6 enabled, whereby approximately 1% of its traffic is native IPv6.

How much and which efforts are required to adopt IPv6 differ amongst actors and depend on each individual case. Therefore, it is practically impossible to reliably estimate the aggregated costs to introduce IPv6 globally. Experience and learning from projects have shown that costs can be kept under control when deployment is gradual and planned ahead. It is recommended that IPv6 be introduced step-by-step, possibly in connection with hardware and software upgrades, organizational changes, and training measures (at first glance unrelated to IPv6). This requires a general awareness within the organization in order to not miss those synergies. The costs will be significantly higher when IPv6 is introduced as a separate project and under time constraints.

Introduction of IPv6 will take place alongside the existing IPv4 networks. Standards and technology allow for a steady incremental adoption of IPv6 by the various stakeholders that will help to keep costs under control. Users can use IPv6 applications and generate IPv6 traffic without waiting for their ISP to offer IPv6 connectivity. ISPs can increase their IPv6 capability and offer this in line with perceived demand.