What are some osquery flags that folks use to limit osquery’s potential impact on systems besides watchdog-specific flags (`watchdog_level`/ `watchdog_memory_limit`/`watchdog_utilization_limit`)? We benchmark our queries w/

profile.py

, test new queries in staging, and set what we believe to be reasonable configurations w/ watchdog settings and such, but I’m looking for some additional insight into how other orgs have osquery configured. Are there any tables that folks disable for perf reasons?