Multiuser Testing

I have a vision in my mind about performance and multiuser testing. The vision certainly applies when test automation is being used, but also when manual multiuser testing takes place. I envision multiple test automation scripts running at the same time, each script simulating multiple users performing a set of activities. I see each script running like a bar of color, like radio frequency waves. The scripts run like music, in a crescendo, the peaks coming close together, almost colliding and creating spikes, and at times running wildly with different peaks and lulls in a cadence of their own. I think of it especially when I hear full symphonic pieces of music, the sense of harmony, of working together.

Each instrument adds to the overall symphonic effect, just as each manual tester adds to the overall impact on the system under test. Even a single manual tester alters the environment by simply being on the system. It takes the entire orchestra to achieve some sounds, just as it takes a multitude of scripts or manual testers to achieve a production load simulation. The activity is no longer about one tester, but rather is about what is achieved as a collective.

Multiuser testing is not the same as performance testing per se. Performance testing is often focused primarily on transaction timings. In the case of multiuser testing, our goal was not about timings, but instead focused on what happens when multiple people execute a specific activity at the same time. Overall, ...

Get Beautiful Testing now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.