Data providers typically deliver data in formats such as parquet, JSON, or CSV. In traditional data warehouse scenarios, these files often land in a persistent staging area. Many organize these stag...
I agree with folks; this is a must-have feature in a lakehouse: the ability to easily separate the physical layer from the abstract layer without constantly copying or materializing data.
Use case examples:
- Rapidly profile and explore raw data (CSV, JSON, etc.) in the lakehouse/files using T-SQL/SQL OPENROWSET in SQL ENDPOINT.
- MVVM style design pattern: Implement an interface (view) on top of raw data files (CSV, JSON, etc.) in SQL ENDPOINT using OPENROWSET to facilitate downstream data flows/ETL
- de-serialize and standardize json schema easily in a view using OPENROWSET and OPENJSON
- expose data in a "fast track mode" to consumer
So basically, just reply with the Synapse Serverless feature. ;-)
Furthermore, it says that this idea is in "Planned" status, so can we get an ETA on it? This information could be very helpful in preparing our migration plan from Synapse to Fabric (or not).
Thanks! :-)
