You are probably already convinced that choosing the right directory server and client software is important. Anyone who has spent any amount of time working with computer systems knows that software often makes all the difference. Let's take a brief look at why choosing the right software for your directory service is so important. Next to developing a good directory design (which we have already discussed extensively, in Part II, Designing Your Directory Service), choosing good software does the most to ensure the success of your directory deployment. A bad directory design can't be repaired by good directory software, but a good design can be ruined by bad software. By "bad software" we mean primarily software that does not meet your needs, although you should also avoid software that is buggy , is poorly supported, or has another serious defect. There is no one- size -fits-all directory solution; directory software varies widely in its capabilities. This is a good thing. Although your directory requirements may overlap greatly with those of other organizations, nearly every deployment has some unique requirements. The greatest challenge in choosing directory software is to find the best match between your needs and the capabilities of the available products. (Directory service requirements were introduced in Chapter 6, Defining Your Directory Needs, and were discussed throughout Part II.) You may run into serious and expensive problems if you do not realize right away that you chose the wrong software. The situation is most common with a growing directory service in which your performance, scalability, and applications needs grow steadily over time. In the worst case, you may not notice that you have a problem until it is too late to fix it gracefully, and the result may be dissatisfaction with your service. For example, if the use of your directory increases over time, you may notice only a gradual degradation in performance. If gradual performance loss is acceptable, there is no problem. However, if your directory server software reaches a cliff at some point and performance falls off abruptly instead of degrading gracefully, some directory-enabled applications may be unable to cope. If applications just stop working one day (which is possible), you will find yourself really scrambling to devise a quick fix. Many hidden costs may be associated with choosing the wrong software as well. Small deficiencies in performance, scalability, ease of administration, and quality of support for applications and standards can lead to increased costs for maintaining your service and all the applications that surround it. Conversely, a directory server product that can accommodate 10 percent more traffic may allow you to deploy five servers rather than six, which could result in a savings of hundreds of thousands of dollars over the lifetime of your service. Another factor that makes choosing the right software crucial is that it can be expensive and time-consuming to replace one software product with another, even though open standards such as LDAP tend to reduce such costs. Some costs you may still incur when replacing directory services software include the following:
As you can see, this is a pretty scary list. Therefore, it is important to perform a proper evaluation of the available software products to make the best choice up front. |