I am looking for some advice on architecture for an app-owns-data scenario embedded report.
We have reports with 8 tables. Four tables are populated directly with data from APIs (2m rows, 2m rows, 10m rows, 100m rows) and we estimate they will increase by 20% a year. In the other four tables we run scripts which perform analyses to write metrics into much smaller tables (1k rows to 1m rows).
I know we would require an Embedded Node such as A1 for capacity but want to understand which other services we need and what exactly each does (e.g. SQL Database, Blob Storage, etc.).
This would be much appreciated and help give us direction before we speak to a data architect in detail.