AWS is a cloud computing platform offered by Amazon. VPT leverages multiple AWS services to provide the ability to rapidly scale and distribute performance tests over multiple geographic regions.

A GUI refers to the graphical windows which a user interacts with by clicking and typing. A web application can have a GUI which is accessed through a web browser.

A key performance indicator (KPI) that broadly refers to the number of requests that your scenario is sending to the target application per second of a test.

The number of times that a user or thread group will repeat during a test. If iterations are specified, then a test will typically end once all the thread groups and users have completed the configured iterations. If the iterations are set to “forever”, then the thread groups and users will repeat indefinitely until the configured duration has been reached.

JMeter is an industry-standard, Java-based, open-source performance testing tool maintained by the Apache Foundation.

A JMeter test plan is saved as a file with the .JMX file extension. It’s an XML file that represents the scenario defined in JMeter. These files need to be uploaded to VPT to run tests.

A key performance indicator is a performance measurement used to evaluate certain aspects of a system. Common KPI include average response time, memory usage, throughput, etc.

Load generators are the servers which run JMeter as well as other VPT services. Each load generator runs independently and allows for scaling tests. Tests with many virtual users are typically run on multiple load generators and can span multiple geographic regions.

The unit of measure used to track test usage in VPT. A load hour represents usage of a single load generator for a single hour. For example, if you run a test with five load generators for two hours, the test will consume 10 load hours (5 load generators X 2 hours). See our Pricing page for details about how many load hours are available for each subscription tier.

The amount of time it will take all virtual users to start up for a test. See JMeter Ramp-up Period.

A service-level agreement defines the level of service and performance that should be expected from an application or service. For example, your application may be required to serve 90% of web pages or API responses within 4 seconds of the request to maintain an agreed upon level of service to your clients.

The application or endpoint that a test is configured to send traffic to. All metrics shown in VPT are calculated based on how the target application responds to the communication.

In JMeter, a thread group is a pool of users that will execute a particular scenario as part of your test. All virtual users within a thread group will execute the same actions. You can create test plans that contain multiple thread groups for more complex scenario execution. See JMeter Thread Group.

A user path is how VPT presents JMeter thread groups when configuring a test.

A virtual user is a single thread that is run in JMeter that executes actions defined in the test plan. Performance tests are configured to run with many virtual users concurrently, to simulate real world scenarios with multiple clients using an application at once.