Directory Structure

Directory information is stored in a data store. A server can contain more than one data store. A data store is the physical location where a naming context is held, and identified by the distinguished name of the naming context it stores. A data store can hold more than one naming context and can have access restrictions as those defined previously.

Information in the directory is organized in a tree structure, called the Directory Information Tree (DIT). The structure of a DIT usually reflects very closely the structure of the information it contains. For example, a directory containing entries for people in a corporation can be organized by division or by location. In general, DIT structures are organizational, geographical, or include both organizational and geographical factors.

When dividing the DIT into data stores to be held on individual servers, you need to take account of the following:

A server can hold more than one data store, with certain limitations:

You cannot search more than one data store at a time.

Alias definitions are local to a data store.

A server can hold both master and replica naming contexts, provided they are in separate data stores. See "Replication" for information about defining a replication strategy.

Example: The XYZ Corporation

Let's assume a scenario in which the XYZ Corporation is a pharmaceutical company, with headquarters in Boston, USA. They have two manufacturing operations, one in San Francisco and one in Paris, and three distribution centers, in Atlanta, London, and Tokyo. There are two research groups, in London and San Francisco, located with the other XYZ divisions in those cities. The Sales organization has three divisions: Europe, US&C (US and Canada), and the Rest of the World (RoW). FIGURE 4-3 shows the functional structure of XYZ Corporation.

FIGURE  4-3 Functional Structure of XYZ Corporation

FIGURE 4-4 shows the geographical structure of XYZ Corporation.

FIGURE  4-4 Geographical Structure of XYZ Corporation

As is common with many organizations, neither an organizational DIT structure nor a functional DIT structure completely meets the directory structure needs of XYZ Corporation, so the network management team decides to combine functional and geographical factors, and to take into account the different usage patterns within the different departments. The result is the DIT structure shown in FIGURE 4-5.

FIGURE  4-5 DIT Structure for XYZ Corporation

In this DIT structure there are ten naming contexts. Each naming context contains entries that are related to a particular geographical or functional area. Each naming context can be stored on a different host, which, given that much of the enquiry traffic is expected to be local to a server, reduces the network traffic. TABLE 4-3 lists the distinguished names of the naming contexts in the DIT structure, and the name of the server holding each data store.

TABLE  4-3   XYZ Corporation Naming Contexts

Naming Context	Server
ou=Boston, o=XYZ, c=US	boston
ou=US-Sales, o=XYZ, c=US	ussales
ou=Atlanta-Dist, o=XYZ, c=US	atlanta
ou=San-Francisco, o=XYZ, c=US	sanfran
ou=Euro-Sales, o=XYZ, c=US	eursales
ou=London-Dist, o=XYZ, c=US	london
ou=London-RD, o=XYZ, c=US	lonres
ou=Paris-Man, o=XYZ, c=US	paris
ou=RoW-Sales, o=XYZ, c=US	rowsales
ou=Tokyo, o=XYZ, c=US	tokyo

A referral system ensures that if an entry cannot be found locally, the directory server can pass the request to another directory server.

TABLE 4-4 shows the referrals defined on each server.

TABLE  4-4   XYZ Corporation Referrals

Server	Referral
boston	none
ussales	boston
atlanta	ussales
sanfran	ussales
eursales	boston
london	eursales
lonres	eursales
paris	eursales
rowsales	boston
tokyo	rowsales