kennyj wrote:Does each BU in a loading run operate independently of all the others? Is there a way to have two or more cooperate?
Yes, that is the basic idea of a "browser user" or BU: each instance is a separate, independent and self-contained browser that plays back its script automatically.
There is no requirement that, in a scenario that is running 100's or 1,000's of BUs, that they all run the same script. In fact, all of them could be running different scripts.
The
second part of your question is somewhat more complicated. What you are asking is, how do you simulate, in an automated testing situation, the
interaction between two [or more] BUs that are running at the same time?
Of course that is possible, and in general the way to think about it is that the two process have to somehow "signal" or "message" each other in a cooperative way. That brings up the issue of intra-process synchronization: if the two processes have to cooperate then they must maintain synchronized communication.
With eValid this is relatively easy to do, given that eValid has so many alternative ways to "synchronize" playback. While most of the synchronization methods in eValid are oriented to waiting (holding back playback) until something happends from the server, that is
NOT always necessary. You can synchronize on the content of a file or on part of an image or anything that eValid can see to synchronize upon.
The simplest way to do this is to have Process "A" send a signal to Process "B" by modifying a web page that Process "B" is synchronizing on. In turn, then, Process "B" can signal to Process "A" by doing the same.
We use this same technique, for example, in coordinating the start of multiple-browser load test scenarios. It works very well!
eValid Support