Ok I got the point and thank you for your respond.

So here is how my data is organised

• I have 2 tables in Snowflake
  • table1 : weekly table containing all the good data
  • table2 : table contains only 1 week of logs for the changes that happened in the first data (updates, deletes, ...)

I should be working with the table1, but as it grows fast and I can't always load it into databricks anytime in a materialised table, the idea were

• to load the table1 once in databricks
• use the table2 every week to update the table1 by using CDC with dlt

What do you think can be the best approach in this case if we are working with dlt?