Automated tests created in Test Modeller are executed on an execution server, either locally or remotely. This server can be created on any Windows machine that is accessible, for instance on your own machine, on a machine in your network, or as a virtual machine.
A remote server is generally preferable, as this means that you can submit execution jobs that run remotely, off screen. You can then carry on working on your machine while the tests execute. If you are working in a team, a remote server is especially preferable. If you set up a shared server on your local machine, the tests will execute on your machine whenever a member of your team submits a job.
Automated test jobs are run using a job engine that runs on the execution server. The server and job engine do not need to be switched on to fire off automation jobs, which will queue up when either or both are switched off. The jobs will then execute when the job engine is next turned on.
Two tasks must therefore be performed before automated test jobs can be executed:
The server must be set up. This is a one off task for each server.
The job engine must be switched on. This is required every time jobs are going to be executed. The server and job engine can be left on indefinitely, but this can be resource-intensive and costly. Jobs can alternatively be queued up while the server is switched off.
Both tasks are explained in this section of the Knowledge Base.
Curiosity also offer a number of ‘Quickstart’ automation frameworks that allow users to execute automated tests with very little setup. The available frameworks here include, Java, REST Assured and C#, with automated tests executing on a server maintained by Curiosity.