How (Not) to Do Load Testing

The results of load testing should always be viewed critically. Let's see this in a few examples, when the numbers lie, and certain factors nullify the efforts of load testing. Tools such as Apache JMeter and Gatling will help us in this. For this reason, some parts of the talk will be Java/JVM-specific.

How many RPS will our service sustain? What did we just test? Why are the numbers lower in the production environment? Why are they higher sale? What if we would test the production environment itself? If you already know the correct answers to these questions, then this is probably not the talk you are looking for. But if you are just getting acquainted with the topic of performance testing or the results of such testing leave you with questions, this is the very case when you can find out about some traps in advance.