Responsiveness as Part of Performance

Java^TM Look and Feel Design Guidelines: Advanced Topics > Part II: Special Topics > 6: Responsiveness > Responsiveness as Part of Performance

Responsiveness as Part of Performance

In Java Platform Performance, coauthors Steve Wilson and Jeff Kesselman describe performance as a general term for several related measurements, among them:

Computational performance
Scalability
Perceived performance, which this chapter calls "responsiveness"

Computational Performance

Computational performance--what software engineers usually mean by "performance"--focuses on fast algorithms, efficient data structures, and economical use of processor time.

Responsiveness and computational performance are not always related. Improving the computational performance of an application improves its responsiveness only if users prefer the change or if the change increases users' speed or accuracy.

It is sometimes possible to improve an application's responsiveness without speeding up the application's code. For tips on how to make such improvements, see Responding to User Requests.

Scalability

The term "scalability" has two meanings:

In the context of computational performance, scalability is an application's ability to perform under heavy loads--for example, large numbers of concurrent users.
In the context of user interfaces, scalability is the ability of a user interface to remain responsive as a user:

Does increasingly complex work.
Tries to gain access to increasing numbers of interface objects--for example, file folders or device descriptions.

Without a scalable user interface and scalability in performance, an application can quickly fall from being highly responsive to being extremely unresponsive.

Perceived Performance, or Responsiveness

Perceived performance, or responsiveness, is based on how fast an application seems to its users--how well it responds to them, not necessarily how fast it fulfills their requests.