1. we do host.each, not sure if we can avoid it, but we could try to loop in batches and load associated data.
2. the status calculation is done per host, which might not be avoidable, but I would like to see if we can wrap it in a host collection instead of a single host to avoid many n+1 queries.

I tried to fix some (n+1) queries and do the processing in batches (batch size didn't seem to have much effect), I don't see any major improvement. On the other hand, my DB is quite small.