How can we use AppMon to find the cause of a performance regression or degraded test?
Hi,
We have a large suite of automated UI-driven regression tests (implemented using Selenium) and we would like to use these tests in combination with Dynatrace AppMon to identify performance regressions in the application under test.
We are using the UEM JavaScript Agent to capture metrics, and we record test runs using the session storage component so we also have purepath data. The problem we have is that it seems to be very difficult to use AppMon to "drill down" and identify exactly where the performance regression is occurring in a test. We would like to be able to perform a simple side-by-side comparison of two executions of a specific test case, so we can tell our developers (for example) “the loading of page X is 3 seconds slower than last week”, or “database query Y is taking 4 s longer than last week”. The best solution we have right now is to filter the purepath comparison dashlet to show only the 2 visits that we care about, but as a test case can have multiple purepaths, we still have to manually find the degraded purepaths. Are we missing something or is this kind of comparison not available?
There is no single way to do this. You can create business transactions on top of your main transactions, chart response times and failure rated and allow comparison here (for example). Such comparison is possible for both UEM and server side perspectives.
Sebastian
