Chapter 7. Looking at the “Abilities”
IN THIS CHAPTER
When you deploy a Web Service, there are a number of things you can do to make it successful. A lot of those items have to do with designing for a set of specific targets referred to as the “abilities.”
Availability— The goal of availability design is avoiding a single point of failure. We will look at how to build your Web Service so that another box could easily take over if another one failed.
Scalability— You want your Web Service to handle more concurrent users as it gains popularity. One of the worst things that can happen to a Web Service is for it to fail because it became too popular and could not handle all the users.
Manageability— When deploying a Web Service, you want to make sure that your systems operations people can easily monitor what is happening inside the system. We will discuss things you can do to make system monitoring easier.
Performance— Of the four “abilities,” this one does not follow the naming pattern. Nevertheless, you need to design things around performance. You need to define minimum response times under given user loads so that you know how fast things ought to perform. Because every application is different, we will discuss deployment options that will make your Web Services more responsive.
Much like Chapter 6, “Security Issues with Web Services,” this chapter should help you think about what you need to do as a developer to make your Web Service perform well on the Internet. One thing that can cause a Web Service (or any Internet-based application) to fail is unanticipated success. By thinking about success early in the design and coding process, you can make sure that the Web Service is ready to handle success without adding much cost. With any luck, you will save time rewriting the Web Service too. The things you will do for Web Services are not much different from things you would do to make a Web ready to handle the “abilities.”