Simple Jmeter JTL Parser
Needs python 3, most likely it is already there, if not see https://docs.python-guide.org/starting/install3/linux/
install python packages via pip3 install pandas, seaborn, matplotlib
To run the command, give a root directory. Have JTLs in root/XX/yy.jtl and XX will be used as experiment name hopefully this is how you have the data already
script will find all .jtl file parse them prints results against experiment name of each JTL
(e.g. output) exprement throughput90p_min throughput90p_max latency90p_min latency90p_max 0 JTL 18137.65 18548.616667 2.0 10.0
throughput values are calculated as follows
- calculate throughout per each second
- take the average over 30 moving window
- take 10th percentile and min and 90th percentile as max
10 and 90th percentile of avg throughput of 30second Also it creates plots with throughput latency behavior for each experiment