Select Page

Load Test Results repository…

I’m going to take a look at ( today.  I’ve currently been parsing jmeter result files and throwing the results into a mysql db.  This gives me the luxury of comparing two test runs across two different builds of the target application.
I’m curious to see the out-of-the-box functionality of the jmstats tool.  With some modification, it looks like it could be just what I’ve needed.

Every test run has specific properties and variables that need to be captured and persisted along with the results.  This could include the number of test clients, the number of test targets, and even the version of both the test itself as well as the version (build) of the target app.  Target application specific properties also need to be set like the allocated jvm heap size, the various cache sizes, and the number of db connection pools, among others.  Currently, I don’t see a place for properties like this in the jmstats tool and perhaps this could lead to a project improvement.

Not to mention the fact that we like to store jmeter configurations like rampup period, delay, thread counts, etc into structured and concise fields.  This would allow for more robust analysis of test result deltas instead of scouring through a field called ‘Notes’ for this unstructured information.

Will update the blog once I’m done mucking around 🙂