I am specifically interested in Functions that are consumers of service bus.

Docs say that for target based scaling, the following formula is being used:

N = event source length / target executions per instance

Further, it says:
In this equation, event source length refers to the number of events that must be processed.

Questions:

1. Does event source length corresponds to number of active messages in service bus subscription?

2. If so, do Functions do polling of messageCount API on Service Bus?

3. If so, at what frequency is the polling done?

Furthermore, for incremental scaling it says:
For non-HTTP triggers, new instances are allocated, at most, once every 30 seconds.

Does this mean that it polls Service Bus messageCount API every 30 seconds or does it, in general, have some other mechanism of determining event source length?