Home > Articles

  • Print
  • + Share This
This chapter is from the book

Performance and Scalability

Have you ever gone to a Web site, clicked on a link, and really waited for a response? Of course you have; we all have. It's annoying and frustrating. Worst are those content-laden sites that are meant to be read like newspapers. You want to jump from link to link, but very time you click, you have to wait seconds (not milliseconds) for the page and the ads and the embedded applets to download. You almost begin to hate clicking on a link because you know you will have to wait. You've learned to classify this kind of site as slow.

Then there are sites that are suspiciously slow. In these cases, you have reason to believe that bazillions of people are trying to connect, and this mass, not the technology, is responsible for the slowness. Say you're ordering a book at a site that has just announced a 50%-off sale. Or suppose tickets for a really hot concert have just gone on sale. When you're able to connect, the site seems unresponsive. When it does respond, it crawls. You guess that the site is buckling under the demand caused by the event. You've learned to classify this kind of site as not scalable.

As users, we have learned what poor performance and scalability are because we have experienced them. As engineers, we would like to understand these faults better so that our own users don't experience them. Because that is the focus of this book, let's start our discussion of performance and scalability by defining our terms.

  • + Share This
  • 🔖 Save To Your Account