I am running a jmeter script against database. My queries are producing huge volume of data as response data(nearly 100000 rows , Request count: more than 30000). I am not willing to limit the rows. So my jmeter instance is facing memory issue.
I am currently using 64gb ram and 8vcpu instance. I have also allocated 56gb as heap memory to jmeter. I am not willing to do distributed testing. If I have to increase the instance memory what should I go for? Is there any way to over come this?
Memory issue can have many faces and you need to take a look at jmeter.log file to see what exact error message you're getting.
If it's about Java Heap Space - unfortunately there is no workaround apart from going to distributed testing.
In case you don't need the full ResultSet you can try switching it to Count Records
Alternatively you can consider migrating to JSR223 Sampler with Groovy language, this way you will get full control on how to work/what to do with results (ensure that you're doing the same operations as the application you're trying to mimic). More information: Interacting with a SQL database
