Forum Discussion
Modeling for Multiple Granularities and 1000+ Customers (OneLake layer)
Hi jaryszek,
When organizing data in a shared Lakehouse by tenant_id and date, files are stored in a nested folder structure like:
/tenant_id=<tenant_id>/year=<yyyy>/month=<mm>/day=<dd>/.
Fabric automatically manages file names and partitions when using Spark or Pipelines. You can view the folder structure by selecting “View Files” in the Lakehouse UI. This method enables scalable and efficient management of data across multiple customers and granularities.Load data to Lakehouse using partition - Microsoft Fabric | Microsoft Learn
Thank you.
"Use a single shared Lakehouse to store detailed data, partitioned by tenant_id and date. "
Please describe exactly how files names should look like? Can you please provide patterns?
Hi jaryszek,
When organizing data in a shared Lakehouse by tenant_id and date, files are stored in a nested folder structure like:
/tenant_id=<tenant_id>/year=<yyyy>/month=<mm>/day=<dd>/.
Fabric automatically manages file names and partitions when using Spark or Pipelines. You can view the folder structure by selecting “View Files” in the Lakehouse UI. This method enables scalable and efficient management of data across multiple customers and granularities.
Load data to Lakehouse using partition - Microsoft Fabric | Microsoft Learn
Thank you.