Hello! I have a project for work and part of the tech stack is Camunda 7. One of the pain points is that several of the exposed remote engine endpoints are a bit slow. For example, the engine-rest/process-instance/count endpoint takes anywhere between 1-3 seconds. Starting a process takes anywhere between 8-20 seconds, though I need to determine how much of that is our application layer. For reference, we only have 5 DMN diagrams and 6 BPMN diagrams; the largest of the latter is only about 3000 lines. I doubt it’s the lack of resources, as the remote engine is hosted on two K8S clusters with two pods each, and CPU / memory usage are well below their max allocations.
How long does it usually take for the engine to process HTTP requests? I saw this post Tune Camunda For Better Performance when searching for an answer, but the OP does not use actual time measurements, which I would like to know more about.