0

In the absence of DML (not leveraging Delta Lake as of yet), I'm looking for ways to manage Static/Fixed-Content Dimensions in a Data Lakehouse (i.e. Gender, OrderType, Country).

Ideally the schema and data within these dimensions would be managed by non-technical staff, but at this point, I'm just looking for development patterns to support the concept technically without being able DML. Preferably with history on source (who added 10 rows to this dimension?)

Thank you in advance for any assistance.

Chris Wilson
  • 3
  • 1

1 Answers1

0

The Lake/Lakehouse/Warehouse should not be the system-of-record for any of your data. So you have another system where that data is "mastered" and then you copy it. Just like anything else.

The "system-of-record" can be a SharePoint List, an Excel workbook on OneDrive, a Dataverse table, or whatever you find convenient and is accessible to the people managing the data.

David Browne - Microsoft
  • 80,331
  • 6
  • 39
  • 67
  • 1
    Thank you for the reassurances that I'm on the right course. I certainly wanted to move away from the old ways of managing the fixed content dimension members in source-control and executed via DML scripts out of DacPacs. I had a feeling the answer was to treat the data just as another source. I will continue to pursue a method to make store the data and make it manageable outside of Data Team staff. – Chris Wilson Aug 17 '22 at 17:37