Introduction
In this article, i want to focus specifically on Materialized Lake Views (MLVs in order to better understand how they work and where they can be applied.
MLVs are still relatively new in Microsoft Fabric and, at first glance, do not fully align with what Microsoft has strongly promoted so far, namely a strict separation of the individual layers. Nevertheless, they represent a very interesting alternative, especially when it comes to reducing complex ETL processes.
In my projects, I generally follow the Medallion architecture, where I use a separate Lakehouse for each of the Bronze, Silver, and Gold layers. Alternatively, the Gold layer can also be implemented as a Fabric Warehouse.
This is exactly where Materialized Lake Views come into play, as they enable a different approach:
Instead of operating multiple Lakehouses for the individual layers, Bronze, Silver, and Gold can be represented via separate schemas within a single Lakehouse. The individual layers are logically separated while being technically created within the same Lakehouse.