Enterprise Toolbox

SAN MATEO (04/10/2000) - You've most likely uttered an expletive or two while in the midst of grappling with the business impact of a telephone area code split. We all know what happens: We have too many voice lines, data circuits, fax machines, pagers, cell phones, and the like.

Sometimes we have devised ways to share circuits, but the net result is that eventually we run out of available numbers, and the new area codes are necessary to enable growth and expansion. Is there a short-term business impact that is derived from the introduction of new area codes? You bet. But we absorb that impact (along with the expletives) as a necessary offshoot of business expansion.

As you might expect, a similar situation is cropping up with network and Web-related addressing. The number and types of devices have been increasing dramatically.

Nearly a decade ago, the Internet Engineering Task Force (IETF) concluded that the current version of IP -- Version 4 -- was reaching the limits of its addressing capabilities. This is a critical point because IP is used to interconnect companies and people -- both on the internal company network and externally on a global basis.

Yet we still want to power up refrigerators with Web-connected virtual machines, surf the Web in our cars, and send e-mails from any location via wireless devices.

The IETF has been working for quite some time on the next version of IP, known as either IPng (Next Generation) or IPv6 (Version 6). The biggest difference between the current IP version and IPv6 is the expanded address space (Version 4 supports 32-bit addressing, whereas Version 6 expands to a 128-bit addressing scheme).

You might think of the of the move to IPv6 as similar to adopting a new area code, although you won't have to reprint the company letterhead or all those business cards. With IPv6, there will be enough unique addresses for every IP-capable device -- whether inside or outside of your company.

But why should your company care about IPv6? There are several reasons. First, we have been dealing with the shortage of unique IP addresses for some time by using NAT (Network Address Translation) to share a single IP address among multiple hosts.

However, NAT is only a short-term solution. It may provide an intermediate means of addressing, but it makes both internal and external network routing much more complicated than it needs to be. Additionally, NAT does not work well with some applications, such as DNS and group-conferencing software.

By using a unique address for each device, network administration costs could be dramatically reduced. IPv6 also helps reduce network administration tasks further with its support for auto-configuration using the network adapter address and a router prefix. IPv6 promises to speed up performance significantly because it will work with gigabit and terabit routers. IPv4 is not designed to handle millions of packets per second.

Increases in performance could be used to deploy multimedia-intensive business applications that could, in turn, increase the user-friendliness factor and expand corporate revenue streams. IPv6 also offers multicast and "anycast" support for those with high-availability or heavy bandwidth needs.

Those with VPN plans will also find that IPv6 is an improvement over IPv4.

Security and authentication measures that were add-ons in IPv4 are now built in; this should simplify VPN deployments.

Getting to IPv6 won't throw the whole world into chaos, either. Most major operating system and router providers will support IPv6, although they are currently at different stages of support.

IPv6 has provisions that support the coexistence of IPv4 and IPv6. In addition, a set of protocol mechanisms called SIT (Simple Internet Transition) is supplied to aid the migration. More information on IPv6 can be found at several sites, including www.ipv6.org, www.ipv6forum.com, and www.ietf.org/html.charters/ipngwg-charter.html.

I'm a fan of few expletives and keeping things simple whenever possible. I think IPv6 holds the promise of both of these things. What do you think? Write to me (without expletives, please) at maggie_biggs@infoworld.com.

Maggie Biggs is director of the InfoWorld Test Center.

Join the newsletter!

Error: Please check your email address.

More about IETFInternet Engineering Task Force

Show Comments