The Sixth Protocol

The Internet as we know it soon may be going away. Its time to bone up (6bone, that is) on the next-generation Internet protocol IPv6.

BY PAUL ROLICH

Do you ever wish your chosen field were one in which whatever you learned five or 10 or 20 years ago would be all you needed to perform your job? I doubt it. I think one of the reasons we chose to become technologists is we enjoy working on a constantly changing playing field. I suppose I still could write mainframe assembly code, but does that ability have any relevance in a world dominated by Web services and near-universal system connectivity? It definitely is not relevant as a hard skill. It does speak to a certain ability or inclination for a logical thought process.

You probably also did not choose to work for an insurance company because you love insurance. You chose to work in technology because you love the challenges and constant changes in computer science. As we deal with those changes, we continually need to be aware not only what those changes are but how and when we can best incorporate them into our places of business. Some of the challenges we face have hard deadlines (such as the over-hyped Year 2000 issue), and some of them are just out therewaiting for us to embrace them. This month, we will look at one of those changes that is looming on the near horizonthe next-generation Internet protocol IPv6.

Scalability and Ancient Protocols

What we commonly call the Internet is a loosely coupled network based on a number of standards and protocols. We usually refer to the base protocols as TCP/IPas in HTTP over TCP/IP. (This is another advantage of being a technologistgeeks always have had a wealth of brand-new terms with which to confuse the huddled masses.) Anyway, TCP/IP stands for Transmission Control Protocol/Internet Protocol. TCP is the connection-oriented, transport-level protocol used in the TCP/IP suite of communications protocols. IP, or Internet protocol, is the most basic protocol used to communicate on the Internet. An IP number is a numerical address consisting of four numbers separated by periods. Each IP address uniquely identifies a certain computer (or device) on the Internet. IP addresses are translated in domain names by Domain Name Servers (DNS) so that we can type www.nationalunderwriter.com instead of 65.171.14.3.

The Internet protocol we currently use has been around for 25 years. RFC 791 Internet Protocol was published in Septem-ber 1981. It is known as IPv4 and has the distinction of being the only Internet protocol that ever has been in widespread use. IPv4 actually is the first published IP. Previous versions (1 to 3) were part of a larger combined TCP/IP. Addresses in IPv4 are 32 bits long. That (232) address space (4,294,967,296) may have seemed sufficient in the early 80s, but by 1992, it was becoming apparent the exponential growth of both computers and devices connected to the Internet eventually was going to eat up those four billion addresses. For comparison, there are about 6.5 billion people on this planet, so if we each had a computer with an assigned IP, IPv4 would be out of addresses tomorrow.

MAC addresses (a Media Access Control address is the unique physical address of each devices network interface card) have 46 unique bits, which means 70,368,744,177,664 unique IDs are available. Actually, I dont believe MAC addresses are expected to be unique. Like GUIDs, there are so many available that duplication in a given environment is unlikely.

A gNATty Problem

We have been dealing with the disproportion between available unique IP addresses and the number of network and Internet connections with various workarounds such as Network Address Translation (NAT), which is the translation of an IP address within one network to a different IP address within another network. Typically, a company maps its local inside network addresses to one or more global outside IP addresses and unmaps the global IP addresses on incoming packets back into local IP addresses. This helps ensure security, since each outgoing or incoming request must go through a translation process that also offers the opportunity to qualify or authenticate the request or match it to a previous request.

By 1992, the IEFT (Internet Engineer-ing Task Force) was considering a number of different recommendations for developing an Internet protocol for the future. In 1994, the recommendation was made for a protocol known as Next Generation Internet Protocol (IPng), which was assigned the IP version 6 and officially is IPv6. Could we have made this any more confusing? Hang onwhat happened to IPv5? There was an IPv5 specification produced in the late 1980s. However, even within the RFCs defining IPv5, it was referred to and considered as an experimental protocol. It really never was intended for general public consumption. So, in the world of Internet protocols, we have the first (v4) and the second (v6).
So, what is IPv6 all about? There are a half-dozen significant differences between IPv4 and IPv6: Expanded routing and addressing capabilities, improved scalability, simpler header formats, greater flexibility, quality of service enhancements, and improved security and authentication.

The most readily obvious and discussed feature of v6 is its vastly expanded address space. With a 128-bit address, we now have the possibility of 340,282,366, 920,938,463,463,374,607,431,768,211,456 unique IP addresses. This supplies what amounts to a measurably infinite number of addresses, which in turn means we will have the ability to connect virtually any device we want to the Internet. An article in InfoWorld (published Dec. 13, 2004) tells us about cows and taxicabs in Japan that not only have been assigned IPv6 addresses but are hooked up with network devices. The cows location and body temperature are being tracked for health and breeding purposes, and the taxicabs have devices hooked up to their windshield wipers so that dispatchers can determine where the most rain is falling and then send more taxicabs where they are most needed. Hmmm.

Information Overload?

At the risk of sounding a bit curmudgeonlier than usual, I will state I think we have gotten ourselves in a situation where we are using enabling technology to enable ourselves to gather too much information. (This is a slightly different use of the popular TLA, or three-letter acronym, referred to as TMI, or too much information.) What is the purpose of enabling the ability of every machine, device, appliance, and living creature to connect to the Internet? We already are moving on a different front to use RFID (Radio Frequency Identification) to tag everything from cows to dairy products. We are embedding GPS devices in vehicles.
Soon I will be able to purchase a gallon of milk at Wally World and have that purchase tracked by RFID and linked forever to me (through my cash card). The imbedded milk-carton RFID then will communicate with the onboard computers in my car that will track my route and ambient conditions. If I decide to stop at the park on my way home for a jog, I will be warned I have only 27 minutes left to get my newly acquired milk into a refrigerator before risk of spoiling becomes unacceptable. When I finally do unpack and store my groceries, my refrigerator, which is connected to the Internet via IPv6, will transmit that information back to Wally World, which will continue to track the state of the milk and eventually will send me an e-mail reminding me I better hurry back to the store and get some more. Sound ridiculous? Of course, yet the above scenario is very close to reality.
Beyond the frivolous and silly, there also is a real concern we rapidly are approaching the time when we truly do have too much data. There certainly is a threshold level in any information processing system where there is simply too much data available to process it usefully.

The Real Reason

All that aside, there is a valid need for a larger address space for Internet devices. We can think of IPv4 as a protocol that addressed the need to network all computers. That 32-bit address space has served us admirably for the last 25 years. It will continue to fill that need for the immediate future. It is expected existing v4 addresses could last another five years or sountil about 2010. Unfortunately, not all available addresses have been equitably distributed. The vast majority of IP addresses have been allocated to North America and Europe. The Far East was given the leftovers. With Internet use in the Pacific Rim growing faster than anywhere else, it is not surprising that area is running out of available IPs and now is leading the rest of us down the IPv6 road. Japan and other Asian countries already are using IPv6, while North America still is pretty much stuck in IPv4. Fortunately, the v6 protocol is designed for ease of implementation. Most currently used network devices are capable of being upgraded to IPv6 via software of firmware upgrades.

The second reason for an expanded address space simply is we have a need to connect devices other than computers to the Internet. It really is irrelevant where we come down on the issue of Internet-connected home appliances or livestock. These things will take place, and we will need to learn to work in that new world. We must remember while we may have some influence over the business process, we ultimately are technology wizards. If a good business case can be presented for assigning IP addresses to chickens (or remote claims agents), we will be expected to make it happen and to make our existing systems work in that world.

All major vendorsCisco, IBM, Microsoft, Juniper, Hitachi, Sun, etc.already are shipping commercially supported IPv6 products. Apple supports IPv6 in both server and client versions of OSX. Microsoft has v6 support in Windows Server 2003 and XP SP1 as well as other operating systems. Linux has integrated with v6 for some time, and Sun and IBM offer both dedicated v6 and dual v4/v6 support. Top-level domain servers support both v4 and v6.

Tell Me About the 6bone!

The 6bone is the backbone of the IPv6 protocol. It evolved first as an experimental network for testing interconnectivity of IPv6 implementations. There exists a world 6bone composed of all regional IPv6 networks. Most regional 6bones are made possible by using technology that allows v6 protocol to tunnel on existing v4 networks. We increasingly are seeing the creation of dedicated 6bone networks (particularly in Asia).

There is no doubt we all will need to interact with IPv6 in the foreseeable future. Whether we choose to interact by keeping our existing systems to the bitter end while allowing our ISP to perform v6 to v4 translations for us or whether we immediately embrace the technology by exposing a dual IPv6/IPv4 front to the world probably doesnt matter. Just know sooner rather than later you will be operating in an IPv6 world, and you need to start planning now.

Want to continue reading?
Become a Free PropertyCasualty360 Digital Reader

Your access to unlimited PropertyCasualty360 content isn’t changing.
Once you are an ALM digital member, you’ll receive:

  • Breaking insurance news and analysis, on-site and via our newsletters and custom alerts
  • Weekly Insurance Speak podcast featuring exclusive interviews with industry leaders
  • Educational webcasts, white papers, and ebooks from industry thought leaders
  • Critical converage of the employee benefits and financial advisory markets on our other ALM sites, BenefitsPRO and ThinkAdvisor
NOT FOR REPRINT

© 2024 ALM Global, LLC, All Rights Reserved. Request academic re-use from www.copyright.com. All other uses, submit a request to [email protected]. For more information visit Asset & Logo Licensing.