In Fabric notebook, loading a delta table is taking 8 hrs and still not complete.

[pshepunde]

I am trying to load a delta table but it is taking a way too long. More than 8hrs it is running but still not complete. How can I troubleshoot the issue? what kind of techniques can be applied for preformance improvement? please find screen shot of logs

@v-nikhilan-msft

[v-nikhilan-msft]

Hi @pshepunde 
Thanks for using Fabric Community.
Here are some steps you can take to troubleshoot and improve the performance:

1.  Data Volume and Transformation: Check the size of your data being loaded. Large datasets will naturally take longer. If there are complex transformations happening during load that might be causing slowness.
   
   

2. Optimize your Delta Lake table: Microsoft Fabric provides several techniques to optimize Delta Lake tables for maximum performance. One such technique is using V-Order, a write-time optimization to the parquet file format that enables fast reads under the Microsoft Fabric compute engines. You can control the usage of V-Order writes using specific commands.
   Delta Lake table optimization and V-Order - Microsoft Fabric | Microsoft Learn
   
   

3. Table Maintenance: Perform ad-hoc table maintenance using contextual right-click actions in a delta table within the Lakehouse explorer. 
   Use table maintenance feature to manage delta tables in Fabric - Microsoft Fabric | Microsoft Learn
   
   

Hope this helps. Please let me know if you have any further questions.

[pshepunde]

Sure, I will go through all pointers you have shared and reply once issue is resolved. 

Thanks a lot!

[v-nikhilan-msft]

Hi @pshepunde 
We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. Otherwise, will respond back with the more details and we will try to help.
Thanks

[v-nikhilan-msft]

Hi @pshepunde 
We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. Otherwise, will respond back with the more details and we will try to help.
Thanks