Directory Services: Design, Implementation, and Management

by Curtis M. Sawyer and Bob Johnson
Implementing a corporate directory service is rarely as simple as buying a new product, testing it a little bit, and then rolling it out. All the old stuff tends to get in the way. That old stuff is likely to include every e-mail system, directory service, and legacy application in your organization. You can not throw it away and start over, so you have to deal with it. This chapter discusses directory deployment and the integration of legacy systems, as well as the issues an organization must consider when evaluating approaches to enterprise directory services.
Organizations typically have many separate, but related, directories. They also typically have different requirements that each of these directories is attempting to meet. Organizations usually want to create an internal directory service that would aggregate information from several sources and make it available to organizational users and applications. Organizations also typically want to make some data available publicly or to a select number of "trading partners."
This situation raises a question: Are these separate directory services, or are they one? They contain different groups of directory information, with separate users, operational and access requirements, and security issues. The final determination depends on many factors, such as existing architectures, sunk costs, and security requirements. Metadirectory vendors these days are not consolidating disparate directories into a single, physical directory, but rather they are synchronizing directories into a...