Hi!
Thanks so much for this response - sorry I'm so delayed in responding.
I ended up spinning up a dedicated Azure flexible MySQL serverrather than trying to set one up on the cluster machines (too many conflicting dependencies). This worked fine, but I definetly think it's not the most efficient solution. Reading in the dump took an hour, and then reading the tables into Databricks took multiple hours.

I did look into processing the dumps directly in spark, based on [this](https://stackoverflow.com/questions/52108677/how-to-extract-tables-with-data-from-sql-dumps-using-sp... thread, but it seemed a little less than reliable for automated tasks. 

I'm going to have to revisit this because I'll have to do some kind of change capture/diff from each weekly dump that ideally doesn't involve reprocessing the entire table, but that's a bit further down on my list

Thanks again!