There are different types of performance testing to measure various risks:
-
Ping tests measure variability in reaching the server so that spikes in network or server availability will not confound performance test results.
-
Speed tests involve single users to identify response times from a range of app functionality.
-
Concurrency tests identify the impact of multiple requests for the same resources at the same time.
-
Load tests stop testing at a targeted load level (for certification).
-
Stress tests increase users (load) to identify the bottleneck limiting further capacity.
-
Scalability tests measure how much capacity really improves when additional/less hardware is added/removed.
-
Endurance tests measure gradual useage of resources (memory, disk space, etc.) over a longer run times.
-
Baseline tests establish a set of metrics for (all) the tests above as the basis for comparison when configurations are changed and another set of tests are run.