The Run Control screen is where you set and schedule new benchmark runs. A benchmark run can consist of one or more benchmark sets. Each benchmark set can consist of one or more tests.

Templates

Templates let you name and save the settings and configuration for a benchmark run to reuse later. Use templates to define benchmark runs that you intend to use again and again. For example, you can set up a template so that every new endpoint joining the Testdriver service will execute the same standard benchmark run. 

Templates

Select a template from the dropdown.

Load

Load the selected template.

Delete

Delete the selected template.

Save run template

The SAVE RUN button is enabled after you configure your benchmark set(s) and endpoint(s). Give your benchmark run a clear name in the RUN IDENTIFIER field to help you identify it in the TEMPLATES dropdown selector.

Auto-schedule for new clients

When enabled, any new endpoints that connect to the Testdriver service will automatically run the assigned benchmark test(s). The benchmark results will be archived in a set based on the day the test ran.

For example, an endpoint system is added on Monday morning, which triggers the auto-scheduled run template. The benchmark result is saved and the run is marked as completed. Later the same day another endpoint is added. This also triggers the auto-scheduled run template, and the result is saved as part of the same run as the first endpoint. The next day another endpoint is added, triggering the auto-scheduled run, but the result is now saved in a new run.

Run identifier

Enter a name for the test run name. Run identifiers make it easy to inspect all results from the run together.

Run description 

This is a longer text field for describing the test run.

Pass amount

Pass amount sets the number of times to execute the benchmark run.

Schedule later

If enabled, you can set a date and time to start the benchmark run. If both are defined, the RUN BENCHMARKS button will change to SCHEDULE LATER to indicate that the run will not start until the given point in time has passed.

After run

After the benchmark run ends, you can choose to uninstall the benchmarks on the endpoint system, uninstall the benchmarks and the Testdriver client, or leave everything installed.

Benchmark set

A benchmark set is a run using one benchmark product, which may include one or more tests.

Benchmark

Use the dropdown to choose a benchmark available under your license.

Pass amount

Pass amount sets how many times to execute this benchmark set.

Wait before start (seconds)

Use this setting to have the endpoint wait for the set amount of seconds before starting the benchmark set.

Timeout (minutes)

Timeout sets the maximum time allowed for each benchmark to complete. The default value of zero (0) minutes means that there is no timeout period. If a timeout period is set, the endpoint system will reboot if the benchmark does not complete after the time specified. Before rebooting, a screenshot is saved in the following path C:\testdriver\log\screenshot.

Benchmark set — Advanced run settings

Enabling this setting reveals additional advanced settings.

Operating systems

When endpoints with multiple Windows installations (see Multi-OS endpoints) are selected for benchmarking, use this setting to select one or more operating systems for the run.

Command line with settings

Enabling this setting allows you to include custom command line parameters and configuration XML for the set.

Reboot before task

When enabled, the endpoint will be rebooted before each task in the set.

Include demo (3DMark only)

Choose whether to include 3DMark demos in the benchmark set. 3DMark demos do not contribute to the benchmark score.

Add benchmark set

Click the button to add another benchmark set to the run.

Computers

Use this section to choose the endpoints that will execute the benchmark run. This panel shows the endpoints that are connected to the Testdriver Service together with a status indicator. 

  • Busy means the endpoint is currently running a benchmark.
  • Idle means the endpoint has been polling for new tasks to run recently.
  • Offline means the endpoint has not polled for new tasks in the past 2 hours and the system is not running any previously scheduled benchmark runs.

Group

Use this dropdown to select a group of systems. You can create and edit groups on the Settings screen.

Sort by

You can sort endpoints by name, ID, or status.

Name

Define a name for the set of selected endpoints. 

Run Benchmarks / Schedule Later

Once you’ve made your choices, you are ready to start the benchmark run.

If the setup is one you intended to use regularly, you may wish to save it as a template to save time for future runs. Click on the RUN BENCHMARKS button to start the run, or if you choose to schedule the run, click on the SCHEDULE LATER button. The benchmark run will now appear in the Scheduled Runs view at the bottom of the screen.