I have a Karate feature with 7 scenarios that makes a total of 2400 test cases.
A) When I run this feature file, after a few minutes, it quickly takes up the CPU usage (100%) and freezes my laptop (I'm on Windows 10 - Karate 1.4.1). From the console it complains about Java Heap Space. I tried to increase max memory in pom file but it didn't help.
B) So I decided to break the feature into 7 smaller features, each has 100 - 800 test cases tops. This time all 2400 test cases can run successfully. I notice when a feature has been finished, the CPU usage is down for a while, like a cooldown.
Just out of curiosity, what difference does it make between A and B approaches? Thanks in advance .