Exam DP-600 topic 1 question 28 discussion

A. Direct Query "Row-level security only applies to queries on a Warehouse or SQL analytics endpoint in Fabric. Power BI queries on a warehouse in Direct Lake mode will fall back to Direct Query mode to abide by row-level security." https://learn.microsoft.com/en-us/fabric/data-warehouse/row-level-security

Direct lake . When users interact with a report built from a Direct Lake semantic model that uses Delta tables and RLS of the warehouse, the DAX queries will operate in Direct Lake mode. This mode is specifically designed for analyzing large data volumes in Power BI and is based on loading parquet-formatted files directly from a data lake without querying a Lakehouse or Warehouse endpoint. Unlike DirectQuery, there is no translation from DAX to other query languages, and it does not execute queries on other database systems. This results in performance similar to import mode, with the added benefit of picking up any changes at the data source as they occur12. Direct Lake mode supports row-level security (RLS), ensuring that users only see the data they have permission to view. It combines the advantages of both DirectQuery and import modes while avoiding their disadvantages, making it an ideal choice for very large models and models with frequent updates at the data source2.

Based on the table in latest documentation Semantic model with RLS supports Direct Lake mode https://learn.microsoft.com/en-us/fabric/get-started/direct-lake-overview

https://learn.microsoft.com/en-us/fabric/get-started/direct-lake-overview

Falls back to direct query

confirm

Answer from Microsoft Copilot: C. When users interact with a report built from a Direct Lake semantic model that uses Delta tables and RLS, the DAX queries will use Direct Lake mode (Option C) by default. This mode allows the semantic model to read Delta tables directly from OneLake. However, if there are certain conditions like memory pressure or unsupported features at the data source, the model might fall back to DirectQuery mode

The correct answer is A. DirectQuery. When users interact with a report built from a Direct Lake semantic model that uses Delta tables and row-level security (RLS), the DAX queries are executed in DirectQuery mode. This is because RLS requires the queries to be evaluated at the source to enforce security rules dynamically, and Direct Lake does not currently support RLS natively. Therefore, the system falls back to DirectQuery mode to ensure RLS is applied correctly.

IT will fall back to direct query since the RLS is configure on the Data source and not in the semantic model

. DirectQuery

A! Read the limitation of direct lake

RLS is based on warehouse

A query always falls back when the semantic model queries a view in the SQL analytics endpoint, or a table in the SQL analytics endpoint that enforces row-level security (RLS).

RLS on the warehouse will fail back to direct query

When a Direct Lake semantic model is created and used, the DAX queries interact directly with the data in OneLake without importing it into memory (as in Import mode) or querying it from a remote data source (as in DirectQuery mode). The key features of Direct Lake mode include: Direct access to Delta tables stored in OneLake. Low latency querying without requiring data to be loaded into memory first. Integration with features like Row-Level Security (RLS) when configured in the underlying warehouse.

Written in docs itself https://learn.microsoft.com/en-us/fabric/get-started/direct-lake-overview it will fallback to DQ

In the scenario, when users interact with a report built from a Direct Lake semantic model that utilizes Delta tables and row-level security (RLS), the DAX queries will primarily operate in Direct Lake mode. So, the correct answer is C. Direct Lake This mode is designed to provide optimal performance for DAX queries, allowing them to read data directly from the Delta tables. However, if there are any limitations or memory pressures, the model may fall back to DirectQuery mode. But under normal circumstances, you can expect the queries to run in Direct Lake mode, leveraging the benefits of RLS effectively