Stress-endurance test | Capacity testing

100% + 1 testing

The 100% + 1 test strategy is useful when an automated test solution that tests the client GUI directly does not exist, but there is an automation tool that can simulate the client's interactions with the server.

The 100% refers to running the 100% automation bucket that runs against the server directly simulating the GUI interactions but without GUI (since the assumption here is that you do not have a tool to test the GUI in an automated mechanism). The " refers to the 1 (or more) testers manually running the test case using the client GUI.

The key benefit of this approach versus the single user functional test is that you test the client GUI behavior while the server is under load. Using this methodology:

• Any server-side defect could be found since the 100% automation bucket is still being executed.

• Any client GUI-side defect that always occurs (either due to basic functional issues or load on the site) could be found since the manual test execution will always encounter it.

• Any client GUI-side defect that does not always occur (say, it occurs on 10% of the times) may not always be found since the manual test execution would need to be done 10 times to detect the defect once, and this may not be enough to generate a detectable error rate, even when that happens.

Although this process is not foolproof, it is probably the next best alternative to using an automation tool that works with the client GUI.