Hello everyone,
I am looking to better understand the back-end processes in which Tableau imports data from live connections and data extracts, but have not been able to find much documentation. My current situation involves a 600 million record table with 30 columns. How would Tableau handle the import of such a large table? What are all of Tableau's back-end steps?
A few key questions. Please do not limit your response to just these, as I am really trying to understand what Tableau is doing behind the scenes.
- At what point in development does Tableau begin reading data?
- Does it read all data once the connection is made, a sample of the data?
- Does Tableau only read data when required (when a dimension or measure is used)?
- What are all the instances in which a re-query is required (filter change, action, etc...)
- Does the re-query look at all the data or only relevant fields?
Any extra information or documentation is greatly appreciated!
Thank you,
Aaron

Hey Aaron,
When I was trying to answer this question myself, I found this 2 part Video series to be really beneficial. It goes over a wide range of Tableau Architecture processes, including how data is loaded in and out.
Introduction to Tableau Architecture - YouTube
Hope this is helpful.
Regards,
Rody