...
Prerequisites for Kafka/batch scaling?
Your workflow must be designed in a way that can process batch workflows which means it has to have at least one unique identifier that is consistent and links individual records such as a user ID, session ID, or other ID field. This field You must have data that is (partitionalble) i.e. that your data can be split and processed independently using one or more fields as partitioning keys. These fields will then be used by Kafka to determine which partitions the UDRs are assigned to.
...