We are using atlas trigger to compute some denormalization. We already used some match to filter events BUT triggers are not scaling up when our cluster is scaling.
Just to be clear: if we heavily increase our workload we see an increase in execution time… any idea how we can scale ?