Examples of Namespaces

Now that we've described the purposes of a namespace and the major decisions that must be made when designing one, we turn our attention to some examples that should help illustrate the points we've been discussing. Here we consider both flat and hierarchical namespaces. (More design examples can be found in Part VI, Case Studies.)

Flat Namespace Examples

We advised you to design your namespace to be as flat as possible within the constraints created by your replication, access control, and other needs. This section describes two examples of flat namespaces for the fictitious company named Example Electronics.

In the first example, a completely flat space is created. For simplicity, the naming attribute is cn . Each entry's name is a sequentially generated number that has no other significance. Figure 9.11 shows the resulting namespace.

Figure 9.11. A Completely Flat Namespace

Our second flat namespace example introduces a bit of hierarchy, with a flat namespace beneath each hierarchical component. The small amount of hierarchy in this namespace is used only to group different types of objects under a common portion of the tree. This arrangement has the effect of insulating namespace changes in one space from the others. It also makes it possible to partition and therefore replicate information on these same boundaries. Administrative control can also be distributed on these boundaries. Figure 9.12 shows the resulting namespace.

Figure 9.12. A Flat Namespace with Some Hierarchy

In our experience, a namespace like the one shown in Figure 9.12 is a good choice for smaller directory deployments. It has enough hierarchy not to hinder you when your needs change (for example, in replication or access control); however, the hierarchy it does have is not based on information likely to change (for example, geographical or organizational information).

Hierarchical Namespace Examples

You may need a hierarchical namespace because of limitations in the directory software you've chosen , constraints imposed on your directory by applications, or other reasons. This section illustrates two examples of hierarchical namespaces based on different hierarchical schemes. Again, we use the fictitious company Example Electronics to illustrate our design.

For the first hierarchical example, assume that Example Electronics is a global company distributed across three continents. It has an office on each continent that must have control over its own data, yet the sum total of the data must appear to the outside world as a single, coordinated information tree.

Taking a lesson from the flat namespace example, the Example Electronics directory designer decides to use a similar namespace in each of its divisions. Because divisions need autonomy in their ability to manage, access, and change their own data, the top-level namespace hierarchy is divided according to a geographical scheme. Figure 9.13 shows the resulting namespace.

Figure 9.13. A Hierarchical Namespace Based on a Geographical Scheme

Our second example of a hierarchical namespace is based on Example Electronics' internal organizational structure and uses the X.500 naming model. Although you might be able to rationalize building such a namespace, we don't recommend it. But in this example the misguided directory designer has decided to forge ahead. To reflect the organizational hierarchy, he's chosen to use the ou attribute with values corresponding to department names . To name leaf-level entries, the cn attribute has been chosen with a value of first name and last name. In cases of name collisions, a number is appended to entries that collide. Figure 9.14 shows the resulting namespace.

Figure 9.14. A Hierarchical Namespace Based on Organizational Structure

The examples presented in this chapter, combined with the guiding principles we've discussed, should help you arrive at the best possible design for your directory's namespace.