Minor The same stream processor can be used with different log configuration in different nodes.
A use case for this is to have a Nuxeo node processing the async work from a mass import and another Nuxeo node processing the async work coming from the Web UI.
The StreamWorkManager will have a different Kafka Log configuration on each node so they are using different topic names.
The Kafka consumer group is not prefixed, this does not prevent the nodes to work as expected.
But it will be better to not share the group name to prevent irrelevant balancing or to prevent blocking of adminitration operation (move consumer positions of one node while another node is running)
So the consumer group should use the same prefix as the log name.
