Free DP-600 flashcards

To reach an Azure SQL Database with public access disabled, reachable only via a private endpoint inside an Azure virtual network, which Fabric gateway gives the least operational overhead?

A virtual network data gateway. It is provisioned and patched by Microsoft and injects into the customer's Azure virtual network, so it reaches a private-endpoint source without anyone running or maintaining a gateway host machine. That managed nature makes it the lowest-overhead fit for a network-bound Azure source.

Common misconceptionThat a firewall service-tag exception lets a cloud connection reach a database with public access disabled. There is no public endpoint for it to route to.

To ingest from an on-premises SQL Server with no public endpoint into a Fabric Lakehouse via Dataflows Gen2, which component bridges the cloud service to the private source?

An on-premises data gateway installed on a machine inside the corporate network, registered to the tenant and bound to the connection. It runs behind the firewall and relays query and result traffic between the cloud Microsoft Fabric service and a source the service cannot reach directly, so it is mandatory for any on-premises ingestion.

Common misconceptionThat a personal-mode gateway can be shared so a whole team reuses one connection to a local server. Personal mode is single-user only.

In Microsoft Fabric, which is the lowest workspace role that lets someone author reports, semantic models, and a Lakehouse but not manage users or settings?

Contributor. Fabric workspace roles are cumulative: Viewer reads only, Contributor adds full create and edit rights for all item types, Member adds sharing and user management, and Admin adds governance. Contributor is therefore the least-privilege tier that grants authoring while withholding access and settings management.

Why can a Fabric Contributor freely create and edit every item yet be unable to add users or reshare reports externally?

Contributor is an authoring role that grants item create and edit but deliberately omits access management. User management and resharing rights are reserved for the Member and Admin roles, so changing who has access requires one of those higher roles.

Common misconceptionThat Contributor can reshare items they personally created. Resharing needs Member or Admin regardless of who authored the item.

Which semantic model storage mode gives import-grade in-memory speed and near real-time freshness over OneLake Delta tables without copying data or scheduled refresh?

Direct Lake reads Parquet-backed Delta tables straight from OneLake and pages columns into memory at query time. This delivers import-like performance while reflecting source updates with no scheduled refresh and no duplication of the data into the model.

What is the standard way to resolve a genuine many-to-many between two dimensions, such as Product and Supplier, for clean filtering?

Add a bridge table holding the distinct key pairs and relate it one-to-many to each dimension. The bridge sits on the many side of two one-to-many relationships, so filter context flows cleanly from each dimension through it, giving unambiguous slicing.

Common misconceptionThat many-to-many cardinality directly resolves a genuine many-to-many between dimensions. It weakens filtering; a bridge table of key pairs is the supported fix.