Directory Service Components
A directory service compiles information about objects of interest in the world and dispenses that information when given a suitably formulated request. The Yellow Pages are a kind of directory service. A library card catalog is another.
People like to have their information classified for easy retrieval. For instance, the Yellow Pages has categories like "Theaters—Movies" and "Restaurants—Outrageously Overpriced." A library card catalog classifies items into "Books—Fiction," "Books—Nonfiction," "Periodicals," and so forth.
Information needs to be readily accessible, as well. People want one-stop shopping. At the same time, you don't want all the information at a single location. This produces bottlenecks, single points of failure, and turf hassles. For this reason, the information in a directory service needs to be distributed among many sources. Each source of information is responsible for maintaining its little piece of the distributed database.
Information needs to follow rules to make it consistent and reliable. Yellow Pages ads contain a limited set of information about businesses in a community. You would not go to the Yellow Pages to look up the current stock price of a company.
A network directory service has entries for users and groups, workstations and servers, policies and scripts, printers and queues, switches and routers, and just about anything else that relates to computing. The attributes for these entries have something to do with their relationship to network services. For example, authentication credentials can be stored in a directory service so users can log on from anywhere the directory service is available. On the other hand, you would not expect to see a user's cologne preference in the directory service.
A directory service is not a general-purpose database. You would not implement a directory service to manage a point-of-sale system in a chain of video stores. But you would consider implementing a directory service to manage the salespeople who log on at the point-of-sale terminals.
Finally, a directory service needs management tools. Administrators need some way to add information to the directory, remove outdated information, and make use of the information that remains. These tools need to be global in scope, straightforward to operate, and aid in diagnosing any problems that might arise.
So let's get down to some basic questions. How does a directory service work? Why does it work that way? How does it break? How is it fixed? And most important, how does it make my job easier so I don't spend all my spare time managing the service that's supposed to be helping me manage the network?