LOL, I can  understand that. The serverless stuff is interesting for some things. Not sure I'd use it for ETL either, but you certainly could just query/update things. Tracking how it works, testing, debugging, all hard things. I don't have a great solution if you want all Azure. Data Factory might be overkill. A VM in the cloud you spin up and down to move things seems like too much as well.

Ultimately you need some process/app that gets data and inserts/updates, right? I'd be tempted to build some process that just grabs data and moves it to some staging on your SQL instance and then another process to merges this into your system. Not sure where the best place to put this is, but maybe there's something on the AWS side that can read and then push to ASD with an ODBC driver.