New Paradigm Resources Group | 30 South Michigan Avenue, Ste. 700 Chicago, IL 60603 | 312.980.7848 | info@nprg.com
April 5, 2007 |
|
IPv6 Are We Taking a Back Seat to the Rest of the World? "I think that IPv6 leadership is a matter of life or death for Americans." (Alex Lightman, CEO, Innofone)
On June 13, 2003, John P. Stenbit, Assistant Secretary of Defense for Networks and Information Integration, briefed Pentagon reporters about the Department of Defense’s adoption of IP Version 6 (IPv6), the next-generation Internet protocol. During the press conference, Stenbit announced plans to switch to IPv6 as the department standard in order to facilitate integration of the essential elements of the Department's Global Information Grid (GIG). Per Stenbit, "Moving to the IPv6 protocol from the current standard of IPv4 would help the department achieve its goal of network-centric warfare and operations by the end of the decade." But the underlying reason for the switch may have been much more fundamental. "The real issue is that end-to-end security is done differently in a network-centered world than it is in telephone communications or in a broadcast, which is basically what we do today," said Stenbit at the time. "Plus, the commercial world is headed to IPv6, and the DOD does not want to be left behind." Stenbit was right about the commercial world heading towards IPv6. But he was wrong if he assumed at the time that American businesses would be leading the way.
The Next Generation Internet Protocol IPv4, which is currently the most widely deployed Internet protocol in use today, has roots which go back to the 1970s. IPv4 is a "best effort" data-oriented protocol that is used on a packet-switched internetwork (e.g. Ethernet). IPv6, developed within the last 10 years, is a large-scale re-design and re-engineering of IPv4 that essentially serves the same functions. However, IPv6 offers critical key upgrades and expanded capabilities in the areas of address space, security, network management, and quality of service. Address Space The most visible – and significant – difference between the two protocols involves address space. IPv4 uses 32-bit (4-byte) addresses which limits the address space to just under 4.3 billion possible unique addresses. In the early stages of the Internet development, this was more than adequate for both public and private Internet users. However, many of these addresses are now being reserved for private networks or multicast addresses. Because of this, there is a very real possibility of an IPv4 address shortage in the not-so-distant future.
In contrast, addresses in IPv6 are 128 bits long versus 32 bits in IPv4. And instead of 4.3 billion possible unique addresses, IPv6 increases that number to a theoretical possibility of about 340 trillion trillion trillion unique addresses. In other words, there’s very little chance of IPv6 address depletion anytime soon. System Management IPv6 advocates like to point out that IPv4 system administration is slow, complex, labor intensive, and error prone. In this area, IPv6 does indeed seem to offer a significant advantage. IPv6 allows for the liberation of routers from fragmentation responsibilities and gives the ability to use the "flow label" rather than more intensive packet inspection. There are immediate benefits to this set up. In resource-constrained environments, IPv6 requires less processing than IPv4, which can result in reduced power demands and latencies, especially for routers. In addition, global routing tables in IPv6 are potentially much simpler than their IPv4 counterparts, something which can further lower memory and computational resource requirements. This is especially significant for mobile applications. Routing for mobile nodes is more efficient in IPv6 than in IPv4, and smooth handover techniques for IPv6 also exist with no IPv4 equivalents. Security Unlike IPv4, security features are standardized and mandated with IPv6. IPv6 offers address manipulation techniques and secure neighbor discovery features that bolster network security. These include resistance to scanning and auto configuration of addresses – both of which makes it much more complicated for a malicious attacker to probe an organization’s systems for weaknesses. Quality of Service Packet fragmentation is a major source of packet delays, or high latency, under IPv4. IPv6 offers numerous improvements in the packet format that help minimize delay, jitter, and packet loss. IPv6 uses a more sophisticated approach to handle data from programs requesting priority handling. The flow label and priority fields in the IPv6 header are used by a host to identify packets that need special handling by IPv6 routers, such as non-default quality of service or "real time" service. More simply put, IPv6-based systems can differentiate between data payloads that are time sensitive, such as streaming video or audio, and those that aren't time-sensitive, such as status reports and file transfers.
Domestic IPv6 – Still in the Trial Phase Despite the advantages and extended capabilities of IPv6, the protocol has had limited deployments within the United States. Apart from the DOD initiative, the most active interest in IPv6 has come from the members of the Internet2 community. The Internet2 community is a consortium of 207 US universities who develop advanced applications and network services across an advanced OC-192c backbone known as the "Abilene Network." The Abilene Network is currently a test bed for IPv6, with the most advanced deployment being the Moonv6 project. The Moonv6 project is a collaborative effort between the Internet2, University of New Hampshire’s InterOperability Laboratory North American, the IPv6 Task Force, and numerous DoD agencies. Currently preparing for its third phase, the Moonv6 project is essentially designed to test and demonstrate the IPv6 technology's effectiveness under large-scale, real-world conditions in the North American market.
But the IPv6 potential has not gone completely unnoticed by the commercial sector. The Internet2 offers "Corporate Partnerships" to private companies who wish to contribute to or participate in the research projects being conducted over the Abilene Network – and IPv6 is currently very "hot" area with a number of Corporate Partners. MCI and Sprint in particular have been deeply involved with IPv6 interoperability testing through the Internet2 community. The two carriers currently only offer IPv6 capabilities on a custom basis in North America, Europe, the Middle East, and Africa; but plans are in place to deliver IPv6 on a much wider scale.
Worldwide IPv6 – Going Commercial, Asia Looms Large In contrast to the U.S., Europe has seen considerable investment from both public and private sectors in IPv6 technology. And there are several commercial and pre-commercial networks and service offerings now. However, the vast majority of large-scale public and private IPv6 deployments are taking place in Asia.
The Japanese government, through the newly created IPv6 Council, is aggressively promoting IPv6 as an integral part of the "U-Japan" (Ubiquitous Japan) initiative. U-Japan is centered on creating a "ubiquitous network society" by 2010 where 100% of all Japanese citizens have high-speed or ultra high-speed Internet connections. The short term initiatives involve encouraging the procurement of IPv6-enabled devices in the government agencies. For the long term, however, the Japanese government has its sights set on IPv6 as an integral component of its national infrastructure. South Korea has paralleled – or even surpassed – Japan in its IPv6 deployment. IPv6 has been deployed since 2004 at several governmental levels including the "e-Government" networks, post offices, universities, local governments, and the Ministry of Defense. The South Korean government has also been collaborating with industry to develop advanced infrastructures such as a broadband convergence network based on IPv6. But South Korea’s commercial sector has been quite active as well. Equipment makers such as Samsung and LG Electronics have been developing IPv6-enabled products and services such as routers and access gateways. On the carrier side, communications providers such as Korea Telecom, Dacom, and Hanaro Telecom are beginning to install gears for upcoming IPv6 service rollouts. Indeed, there are strong indicators that the carriers are quite committed to an IPv6 transition. As an example, Korea Telecom recently secured 17 trillion IPv6 addresses in preparation for the age of "ubiquitous communication."However, China seems to be the one to watch. China has instituted a full adoption policy of IPv6, which is a key element of the China Next Generation Internet (CNGI) program. The CNGI initiative was launched in 2004 by China’s National Development and Reform Commission with the primary purpose of creating a countrywide IPv6 backbone. At the present time, the CNGI infrastructure connects more than 200 universities and 20 cities across China. This is a pure IPv6 network, and the CNGI is making some substantial headway. To give you an idea of what’s possible on this network, the CNGI recently created a 40 Gbps link on a long-haul deployment between Beijing and Tianjin.
Future Implications So if IPv6 is causing such a stir overseas, why is it largely being ignored in the United States? The short answer to this question is that IPv6 has not yet resonated with large enterprises here. While IPv6 offers a number of technological advancements in address space, security, network management, and quality of service, the majority of enterprise customers just simply do not (yet) see the ROI benefits of transitioning to an IPv6-enhanced network. To begin with, legacy technologies – including the current IPv4 standard – will not disappear overnight. Operating a "dual-stack" network will most definitely be more expensive and require much more training and IT staffing, at least in the short term. When you couple this with the fact that only a handful of research institutions are using IPv6 – and that the U.S. government may not be significantly deploying it until 2008 – it’s easy to forget that IPv6 is even a viable option. But U.S. companies that choose this route could be doing so at their peril. On the manufacturing side, U.S. companies such as Cisco and Juniper have significant advantages in the global market because they set the IP-related standards world-wide. But with countries like China embracing IPv6 applications – and domestic competitors like Huawei following suit – it may just be a matter of time before this advantage becomes outmoded. And the global demographics seem to be supporting a transition to IPv6 as well. China has already surpassed the U.S. in both Internet users and total Internet use. We may soon have a situation where China (in conjunction with Japan and South Korea) can reset the standards with regards to all things IP. And that could lead U.S. companies to a situation where they might have to "go the Google route" and kow tow to an already enormous market that just happens to now have 99% of the Internet traffic for the new IP standard. With the upcoming Beijing Summer Olympics, 2008 will undoubtedly be a very big year for China. But there’s another "medal event" taking place right now that involves the very future of the Internet – namely, IPv6 leadership. And unless the U.S. is willing to look at IPv6 much more closely than it has so far, China might just also take the gold in the 2008 Internet Olympics.
|
