Advance your Data & AI career with 50 days of live learning, dataviz contests, hands-on challenges, study groups & certifications and more!
Get registeredGet Fabric Certified for FREE during Fabric Data Days. Don't miss your chance! Learn more
As it sits on top of OneLake like other items (DWH, DL), and OneLake uses parquet files, isn't it a big overhead as parquet files are immutable by design, so Delta tables adds new parquet files with the changeset? How does Fabric SQL DB manage those files? Apply VACUUM Command in the background?
Thanks in advace for your answer.
Solved! Go to Solution.
Hello @saglamtimur
Very interesting question
Microsoft Fabric SQL Database stores operational data in its native Azure SQL Database engine format for transactional processing, while automatically replicating a near real-time copy to OneLake as Delta tables (Delta Parquet format). This dual storage approach balances OLTP efficiency with analytical flexibility.
"The server name is indicated in the red box. This time, it contains database.fabric.microsoft.com, which indicates we’re directly connecting to the database engine and not the delta tables in OneLake. The name of the database (indicated in the green box) is also not the name we specified at creation time. That name is preserved for the SQL Analytics Endpoint. The actual database name is the original name with a GUID appended to it."
this quote from the below blog is absolutely making sense to me
https://www.mssqltips.com/sqlservertip/21367/what-is-sql-database-in-microsoft-fabric/
If you find this information helpful, please give kudos and accept this answer
Ps: I am not able to find any official ms docs for this
Hello @saglamtimur
Very interesting question
Microsoft Fabric SQL Database stores operational data in its native Azure SQL Database engine format for transactional processing, while automatically replicating a near real-time copy to OneLake as Delta tables (Delta Parquet format). This dual storage approach balances OLTP efficiency with analytical flexibility.
"The server name is indicated in the red box. This time, it contains database.fabric.microsoft.com, which indicates we’re directly connecting to the database engine and not the delta tables in OneLake. The name of the database (indicated in the green box) is also not the name we specified at creation time. That name is preserved for the SQL Analytics Endpoint. The actual database name is the original name with a GUID appended to it."
this quote from the below blog is absolutely making sense to me
https://www.mssqltips.com/sqlservertip/21367/what-is-sql-database-in-microsoft-fabric/
If you find this information helpful, please give kudos and accept this answer
Ps: I am not able to find any official ms docs for this
Thanks for your replay.
I read "Microsoft Fabric SQL Database stores operational data in its native Azure SQL Database engine" kind of information somewhere. But, every document I saw about Fabric SQL Database puts it on top of OneLake. So that, I want to verify how it is actually working.
Also, from the screen-shot I see port 1433 is used. So another proof that it is SQL Database engine. And yes, an official MS doc would be perfect.
Thanks again for your clear explanation.
Advance your Data & AI career with 50 days of live learning, contests, hands-on challenges, study groups & certifications and more!
Check out the October 2025 Fabric update to learn about new features.