So as a "for instance", you could query all system processes every 15 minutes or so, and then use a decorator for all of your queries that adds the

host_uuid

and

username

to all results, so every 15 minutes you'd get all processes, but also a "heartbeat" of the host and user