In a report built from a Direct Lake semantic model using row-level security, which query mode is utilized?

The use of row-level security within a Direct Lake semantic model indicates that the architecture is designed to work in real-time with the underlying data while enforcing security policies that restrict access to certain rows based on user permissions. In this context, DirectQuery mode is specifically tailored to support such dynamic queries against large datasets, allowing for the application of row-level security at each query execution.

When employing DirectQuery, Power BI or related reporting tools fetch data directly from the data source as needed. This means that the security measures, including row-level filters, can be applied in real-time, ensuring users only see data they are authorized to view. Consequently, it perfectly aligns with scenarios where data sensitivity and user-specific access are paramount.

In contrast, Import mode would cache data during report initialization and would not apply row-level security dynamically. While Dual mode can support both Import and DirectQuery, it operates under conditions where data is not being filtered on a per-query basis as effectively as DirectQuery. Thus, utilizing Direct Lake with row-level security necessitates the flexibility and real-time capabilities provided by DirectQuery, making it the suitable query mode for this situation.