When exactly and how Data Studio loads data into reports? (rules and in which order)

Question

I am trying to decrease load on database by Data Studio among other BI tools, and need to understand how, in which order and when Data Studio pulls data from data base.

I would like to leverage the wisdom of the crowd to check if I understand it correctly. The only information about it I found in the documentation is here, so overall it is still gray area for me.

Here is how I think it works (all of the points bellow are not applied to extracted data):

1. Data Studio pulls data as defined in data freshness from our connected database, let's assume it does it every 12 hours in our case: refresh is done exactly as defined by filters and controls in each page of the report and not by the actual query in the data source. So if the query has a where clause to pull 12 months of data, but filters and/or controls are set by default to one week, Data Studio will only pull a week of data.

2. When somebody opens the report, data is not pulled from the database itself, but from a data saved in cache when refresh was done (BTW, where is it kept?)

3. When somebody changes controls, data is pulled again, this time from the database itself. Will that new data set be available for another user should he/she changes controls the same way, let's say 2 weeks back, not one week as was defined by the default for that control?

Additional question:

What if I have hundreds of reports all pulling a lot of data? Will Data Studio still keep pulled data from each refresh in cache? There is a limit for queries for each data source of 150,000 rows for Postgres, for example, but when you have hundreds of reports, the volume can be huge. For extracted data the overall limit is 100 MB, so I wonder if there is a limit for refresh cache.