Agentless Scanning Schedules - suggestions or whitepapers

Is there anything published or any ideas in lieu of a document that suggests schedules for agentless scanning of Universal Discovery.  I find that I am constantly balancing IP's on probes and trying to setup the schedules to make the most use of the needed discoveries.  Maybe there's a ratio of Probes to IP's I am not aware of.

In addition, is there a way to make the Host Connection by SNMP run faster.  I would guess that making drastic changes in the queries are the way to go, but just seeing if there are some other methods out there that I may have not considered.

Has anyone come up with a reliable schedule or a way to balance number of IP's against the number of Data Flow Probes needed for a desired frequency?