Queue tests on Automate TurboScale

Queue your tests over and above your Grid concurrency limit for running concurrent tests

The automation grid created on Automate TurboScale supports running the number of parallel tests that you have configured(50 as default). But with test queueing, you can schedule more parallel test requests on the automation grid.

The following are the salient points for queueing your tests:

1. You can launch an additional number of parallel tests with different browser configurations above the concurrency limit that you have configured.
2. The additional tests will be queued and they would run as and when your existing running tests complete their execution.
3. For instance, if you want to run 20 additional tests, apart from your configuration limit of, say, 100 parallel tests, we will queue the additional 20 tests until one of the 20 initial tests finishes, and a slot is available for execution.
4. With queuing, you don’t need to worry about managing your test pipeline - we automatically take care of scheduling and execution for you.

With this feature, the grids will have a max queue length equal to the number of browser concurrency configured.

For example, if you have created a grid with 200 concurrent tests, you can queue up to 200 more tests.

Scenarios where queuing helps

Queuing is enabled by default to ensure you don’t have to worry about managing the test execution on your end. This helps especially in the following scenarios:

• When you have multiple projects / PRs to test using the same Automation Grid. If the other project is already running at full capacity, your new tests are better off in a queue than getting discarded altogether.
• When multiple users in your team use the same grid for their projects. Rationing of parallel tests across teams can be inefficient - instead, queuing can help you manage the test executions more simply.
• If the number of total test cases is within the queue limit and if you don’t want to write some logic to dispatch tests, you can just submit all your tests to the Automation Grid, and it will automatically manage the test runs for you.

In a few rare cases, BrowserStack will automatically put your tests in a queue for a short period so your other tests can run first. This is usually done if the browser & OS configurations you want to run your test on have a long start-time delay (either because of cloud platform restrictions around the number of available worker nodes or booting-up time). The previously queued test will be automatically dequeued and run whenever a slot becomes available.

How does test queuing work

Queues are managed at a Grid level. This approach grants the allotted queues according to Grid configuration, thus letting tests be added to the queue when grid utilization reaches maximum concurrency.