Should you worry about IPv6?

by Stan Schatt

December 15, 2000 —

After the Y2K scare, network managers don't frighten as easily. Take the spate of hysteria regarding IPv6. "We're running out of addresses," cry the alarmists. "You need it for quality of service," say vendors trying to peddle their wares. But so far, there is no wellspring of support for IPv6 among organizations that deploy networking equipment.

In addition to QoS enhancements and more address space, IPv6 offers better multicasting and some improved native security features. Yet many network managers seem content to use workarounds such as network address translation (NAT) and private IP addresses rather than begin the painful process of migration. Microsoft, whose participation is an absolutely critical part of any IPv6 migration, has refused to commit to supporting it in Windows 2000.

Luckily, some of the functionality of IPv6 can be offered without requiring an immediate migration from IPv4. The Internet Engineering Task Force's Realm-Specific Internet Protocol, for example, undertakes some of the security and addressing issues associated with the earlier protocol. RSIP features a replacement for NAT that provides a way to get around NAT's inability to read encrypted packets. RSIP is only in its initial stage of development, however, so it probably won't be available for another a year or so.

Because it will take several years before most devices are equipped with IPv6, now is a good time to start developing a migration strategy to IPv6. One approach would be to consider deploying RSIP within the next couple of years as an interim step until the deployment of IPv6 is widespread enough to make it worthwhile to migrate, or until you have a pressing need for a function that IPv6 facilitates, such as QoS.

ITworld.com