Fabric Warehouse vs Lakehouse

[Rohitb]

I am new to fabric world & we need to mov our current PAAS framework (move away from Azure SQL DB ) into Microsoft Fabric. I see there are 2 options for SQL World in fabric with 1 being delta tables in fabric lakehouse & 1 being fabric warehouse, so on what basis should I make a decision of selecting either lakehouse or warehouse ? 
Is there any guideline that helps to decide between lakehosue or warehouse ?

[NandanHegde · Accepted solution]

And some of the other additinal points of top of what @v-nikhilan-msft mentioned:

1) Fabric warehouse has a DBT adapter , so in case if the org has need for transformations and analytical aspects; one can leverage the DBT aspect

2) Based on my understanding, lakehouse doesnt have any Backup stratgey for DR whereas fabric warehouse has restore points

[v-nikhilan-msft]

Hi @Rohitb 
Thanks for using Fabric Community.
Yes, there are guidelines that can help you decide between a Fabric Lakehouse and a Fabric Warehouse. The choice between the two primarily depends on your team’s skill set and the type of data you’re working with.
Here are some key differences between the two:

Data Types: Lakehouse is designed to handle structured, semi-structured, and unstructured data, while Warehouse is primarily for structured data.

Developer Persona: Lakehouse is typically used by data engineers and data scientists, while Warehouse is used by data warehouse developers and SQL engineers.

Developer Skill Set: If your team is more comfortable with Spark (Scala, PySpark, Spark SQL, R), then Lakehouse would be a better choice. On the other hand, if your team prefers T-SQL, then Warehouse would be more suitable.

Data Organization: In Lakehouse, data is organized in folders and files, databases, and tables. In Warehouse, data is organized in databases, schemas, and tables.

Multi-table Transactions: Warehouse supports multi-table transactions, while Lakehouse does not.

Scalability: Both Lakehouse and Warehouse use Microsoft Fabric’s compute bursting behind the scene to speed up the performance when needed.

SQL Analytics Endpoint: In Fabric, when you create a lakehouse, a SQL Analytics Endpoint is automatically created. The SQL analytics endpoint enables you to query data in the Lakehouse using T-SQL language and TDS protocol.

Fabric decision guide - choose a data store - Microsoft Fabric | Microsoft Learn

Remember, the choice between Lakehouse and Warehouse isn’t necessarily an either/or decision. They can be used together for a best-of-breed analytics strategy.
Better together - the lakehouse and warehouse - Microsoft Fabric | Microsoft Learn

Please refer to these links for more information:
https://blog.fabric.microsoft.com/en-US/blog/lakehouse-vs-data-warehouse-deep-dive-into-use-cases-di...
https://www.advancinganalytics.co.uk/blog/2023/6/13/data-lakehouse-vs-data-warehouse-in-microsoft-fa...
https://www.linkedin.com/pulse/choosing-between-lakehouse-warehose-microsoft-fabric-jovan-popovic/
https://centida.com/blog/articles/warehouse-vs-microsoft-fabric-lakehouse-choosing-best-option/
https://www.linkedin.com/pulse/difference-between-microsoft-fabrics-warehouse-lakehouse-simon-devine...
https://debruyn.dev/2023/fabric-lakehouse-or-data-warehouse/
https://christopherfinlan.com/2023/05/25/lakehouse-or-warehouse-in-microsoft-fabric-which-one-should...


I hope this helps! Let me know if you have any other questions. 

[v-nikhilan-msft]

Hi @Rohitb 
We haven’t heard from you on the last response and was just checking back to see if your query has been resolved. Otherwise, will respond back with the more details and we will try to help.
Thanks