Running the Simulation

Once the template has been finalized, its time to actually run the simulation itself.

Once you have defined a template, it is time to run it. In the simulate section you specify how many clients do you want to simulate and how long and frequent the messages should be sent. Let's have a look at the screenshot.

Here is a quick explanation of the settings

Basic Parameters

Simulation Id: Each run of the simulation is assigned a unique job name, which could be retrieved using JobId() helper API in runtime. The name would be automatically allocated for you, but you could choose to overwrite it to something easier to remember.

Number of Clients: Specify how many total clients you want to launch here. This is limited by your current account settings and can be upgraded based on your license.

Repeat Message: How many messages you would like to send in total throughout this simulation run.

Gap between each message: What should be the interval between sending each message. The minimum value is 1 second and it could be large as well. This will decide the total duration of the simulation.

If you are doing performance testing, inter message gap is the minimum pause duration before a client sends another message. In real world test, due to server and client latencies, the gap may be larger than what is specified. Thankfully, in the result analysis tab, you could see how exactly the mean, minimum and maximum latencies have been measured.

Save State Information: You have an option to capture and see state object information for all clients through all iterations. Collecting this information usually have a slight impact on test performance. If you are running a very large test, this parameter may be automatically disabled.

Scheduling

It is possible to start the simulation right away, at a later time or even create a recurring run. If you have scheduled a simulation in the future, it could be cancelled by going to the Job page.

Simulation could also be launched through the APIs.