Overview

In an industry that moves so fast that, hyperbolically, if you go for lunch, you become lunch, prognosis is a risky business. Yet ATM as an intellectual conception dates from the late 1980s and commercial ATM products have been on the market since 1992. Much has happened and much remains to happen. Although the connective thread may appear somewhat strained, in this last chapter several subjects will be discussed: the ATM Forum s recent Anchorage Accord on new standards, the rise of fast Ethernet and the promise of gigabit Ethernet, the standards process versus the market, and ATM in the longer term.

10.1 THE ANCHORAGE ACCORD

At the ATM Forum s mid-April 1996 meeting, the group agreed, in what has been called the Anchorage Accord or Anchorage Freeze, to a major slowdown in the ATM standards process. Observers have noted that since 1991, the Forum has approved nearly 50 specifications, that 20 more are due by the end of the summer of 1996, and dozens more are in progress. This pace has proven hard on the vendors and hell on the buyers.

The Accord s slowdown involves several sensible courses. First, backward compatibility will be required for all future releases. (UNI 3.1, for example, was not backwardly compatible with UNI 3.0.) Second, it identifies two classes of specifications foundation and expanded features. Foundation specifications, which will be not fully defined until 1997, will include the core LAN and WAN ATM specifications, such as UNI signaling, the private network-node interface (P-NNI), the broadband intercarrier interface...