Reliable with excellent support but needs better source control integration and versioningSource control integration and versioning could be better. Currently, StresStimulus does not natively support any form of change control management. Result analysis (snap-out tabs for side-by-side comparisons) need improvement. Multiple sets of analysis results can be opened in the user interface at the same time, and comparison reports can be generated to compare some aspects of pairs of test runs. However, the interface that displays the multiple sets of results only allows a single result set to be viewed at any one time & uses tabs to switch between the individual result sets. It would be useful to have the ability to view multiple tabs at once, either side by side in the interface or by snapping tabs out of the interface into their own windows to be viewed together with others. Graphing (collect PerfMon data as part of the tests and then have the ability to create different graphs with that data after the test completes to help with analysis) could be better. Graphs and their contents have to be defined before tests are run. Often, as a result of running tests and viewing the results collected, it is necessary to have graphs showing/comparing different metrics collections than those specified before the run. Other tools collect metrics that are not tied to pre-defined graphs & allow the metrics to be used to construct graphs as required by subsequent analysis & investigation. Nested tests/test case groups/test solution files (ability to build a hierarchy of test cases/test case groups/solution files) are required. Within a StresStimulus, tests are organized in just two levels, test cases, and test case groups. As is possible with other tools, it would be helpful to have a hierarchy with multiple levels of nested test cases/test case groups (or even the solution config files themselves). It would make it easier to manage multiple sets, combinations of tests, load profiles & test mixes if a high-level solution file (or test case group) could run a collection of lower-level ones, which in turn end up executing multiple different base-level test case groups and individual test cases.
