Conclusion

This chapter was designed to give you a feel for some of the tools available to you as a Web developer, how to take advantage of these tools, and some of the common problems to look out for. Though much of this chapters advice may have seemed obvious, its important that when you find performance problems in the development cycle, your team develops a structured approach to identifying the causes.

By keeping the information discussed here in the back of your mind, you should be able to shave off a lot of time when developing high-end applications. The chapter stressed the importance of constant measuring to make identifying problems easy.

Although problems are easy to detect and isolate on their own, we all know what its like to make some major revisions to a production application and suddenly find that the application is running abnormally (too slowly, with random crashes, or behaving strangely under load). In these situations it can be extremely difficult to pinpoint the problem unless you take a systematic and constructive approach. This technique is aided not only by the many tools available, but also through the use of constant testing, which allows isolation of the problem (or the cause of the problem) to a reduced number of code paths.

In Chapter 4 we covered some of the basics of your ISAPI DLL and in Chapter 9 we showed how to expose your Perfmon counters as a service from your ISAPI DLL. In the next chapter, we take a closer look at ISAPI services, what services ATL Server provides, what sorts of services you should expose, and how they work.