Hello,
I would like to get your thoughts regarding our following design approach.
1- We have build synapse pipelines
2- Pipelines transfer data from on-prem sql server to Azure blob storage called "Stage1" as .parquet files stored in daily folders
3- Stage2 in the blob storage is populated with full data as a one off load and stored as .parquet files
4- Stage2 is populated daily only with the upserts as .parquet. These are the Delta files and we have set the inline dataset type as Delta
5- In SQL Scripts folder we have scripts to create views, i.e. view1, view2, etc. to read the .Parquet files in Stage2
6- The SQL Scripts are created in the serverless Built-in database and NOT dedicated
7- There is no Workspace being used.
8- Power BI uses the connectionstring to the Built-in database to read the views (Reading .Parquet files) created above.
Q1-What do you think of this design?
Q2- In-particular where power BI reads the views to present data
Thank you