Forum Discussion

InsightsByV's avatar
InsightsByV
Super User
6 months ago
Solved

Issue with Direct Lake Semantic Model with SAP Mirror Database

Hi All,

 

I am currently experimenting and have been stuck, it would be great if anyone can guide me here!

 

What I want to do: 

I have 4 views in my SAP Mirror Database in Fabric. Now I want to use these 4 views and 2 other custom dimension tables (1. Date Dimension , 2. Dim1) and then create a sematic model with Direct Lake mode. I'll be using this to build a Power BI Dashboard in Direct Lake mode.

 

Where Am I stuck:

Case1) SAP Mirror Database won't let me create these 2 custom tables inside of it. 

Case2) If I create a Lakehouse, create these 2 custom tables there and then add shortcuts for these SAP mirrored database - it should logically work but then it will not let me add views so I'll have to use tables from mirrored database instead of views. So, lets say, I created these 4 tables instead of 4 views and use them using shortcuts in my lakehouse, I'll be able to create semantic model out of these.

 

Now my Question:

How can I deal with this scenario? as I do have to use mirrored data in my dashboard along with these 2 custom tables. Will Case2 work? and it will work, will it show mirrored data, properly in my dashboard? created using Direct Lake mode?

 

 

Thank you in advance!

Need Help

5 Replies

  • rohit1991's avatar
    rohit1991
    Super User
    6 months ago

    Hii InsightsByV 

     

    A Fabric SAP Mirrored Database is read-only, so you cannot create custom tables or views inside it, and Direct Lake works only with physical Delta tables, not mirrored views. To use mirrored SAP data together with your own dimensions (Date, Dim1) in a Direct Lake semantic model, you must materialize the mirrored views into Lakehouse tables (using Dataflow Gen2, Notebook, or CTAS), then create your custom dimensions in the same Lakehouse and build the semantic model on top of those tables. This approach is fully supported and the data will still reflect SAP correctly, based on the mirror and refresh frequency.

  • cengizhanarslan's avatar
    cengizhanarslan
    Super User
    6 months ago

    Direct Lake semantic models can only read Delta tables from OneLake (Lakehouse/Warehouse). A Mirrored Database (and especially its views) isn’t the same storage layer for Direct Lake authoring.

     

    • Keep SAP Mirroring as your ingestion/replication layer.

    • Create your Date dimension and Dim1 in a Lakehouse.

    • Materialize what you need from the mirror into Lakehouse tables (Delta) — not views.

     

    • v-menakakota's avatar
      v-menakakota
      Community Support
      6 months ago

      Hi InsightsByV ,
      Thanks for reaching out to the Microsoft fabric community forum. 

       

      I would also take a moment to thank  cengizhanarslan  and rohit1991  , for actively participating in the community forum and for the solutions you’ve been sharing in the community forum. Your contributions make a real difference.

      I hope the above details help you fix the issue. If you still have any questions or need more help, feel free to reach out. We’re always here to support you 

       

       

      Best Regards, 
      Community Support Team