Data Warehousing & Ingestion Pipelines

What term describes the process in which data is transformed before being loaded into an analytical store?

• AELT (extract, load, and transform)
• BETL (extract, transform, and load)✓
• CSchema-on-read
• DMirroring

Answer: In ETL, data is cleaned and restructured before it lands in the store, which is the reverse order to ELT. ELT instead copies data to the store first and transforms it afterwards. Schema-on-read applies structure only when data is queried, and mirroring is Fabric's continuous-replication feature, not a transform-order pattern.

Which Microsoft Fabric component provides the tenant-wide data lake that every Fabric workload shares?

• AFabric Lakehouse
• BFabric Warehouse
• CReal-time hub
• DOneLake✓

Answer: OneLake is the single, tenant-wide storage layer that every Fabric service reads from and writes to directly. Fabric Warehouse and Fabric Lakehouse are analytical store experiences built on top of OneLake, not the storage layer itself. The real-time hub is a catalog for discovering streaming data sources, not a storage layer.

A company wants a fully managed, SQL Server-compatible relational data warehouse that stores its data in OneLake with strong schema enforcement. Which Microsoft Fabric component should it use?

• AFabric Data Factory
• BFabric Warehouse✓
• CFabric Lakehouse
• DDatabricks SQL Warehouse

Answer: Fabric Warehouse is the fully managed, SQL Server-compatible relational data warehouse backed by OneLake, built for structured data and strong schema enforcement. Fabric Lakehouse instead stores data in Delta Lake format and suits mixed or semi-structured data. Fabric Data Factory builds ingestion pipelines rather than storing data, and Databricks SQL Warehouse is part of Azure Databricks, not Microsoft Fabric.

Which Microsoft Fabric feature lets Power BI read Delta tables directly from OneLake without importing or pre-aggregating the data?

• ADataflows Gen2
• BFabric Mirroring
• CA OneLake shortcut
• DDirect Lake mode✓

Answer: Direct Lake mode lets a semantic model query Delta tables in OneLake directly, combining fast in-memory analysis with data-lake scale and no import step. Dataflows Gen2 is a low-code tool for building transformation logic, not a query mode. A OneLake shortcut is a live reference to external storage, and Fabric Mirroring continuously replicates an external database into OneLake — neither is how Power BI reads data at query time.

In a star schema used by a data warehouse, what does a fact table typically store?

• AMeasures and relationships defined in Data Analysis Expressions (DAX)
• BLive references to files stored in external cloud storage
• CAttributes, such as product names and categories, used to group or filter data
• DNumeric values related to one or more dimension tables✓

Answer: A fact table holds the numeric measures being analyzed, such as sales order data, and relates to dimension tables that represent entities like product, store, and time. Attributes such as names and categories belong in dimension tables, not fact tables. DAX defines calculations in a semantic model rather than describing fact-table content, and live references to external files describe a OneLake shortcut instead.

A retail analytics team extends a basic fact-and-dimension schema by linking the Product dimension table to a separate Category table. What is this extended design called?

• AA star schema
• BA snowflake schema✓
• CA lakehouse
• DA semantic model

Answer: When dimension tables are further related to additional detail tables, such as linking Product to Category, the design is called a snowflake schema. A star schema is the simpler design in which dimension tables relate directly to the fact table without extra linked tables. A lakehouse is a hybrid storage architecture, and a semantic model is the tabular layer that defines measures and relationships for reporting, not a warehouse table design.

Which Microsoft Fabric tool provides a low-code, visual way to build reusable data transformation logic using Power Query?

• ADataflows Gen2✓
• BFabric Notebooks
• CEventstream
• DPipelines

Answer: Dataflows Gen2 is the low-code, visual experience within Fabric Data Factory for building reusable transformation logic using Power Query. Pipelines instead orchestrate multi-step data movement and transformation workflows by chaining activities. Eventstream handles real-time streaming ingestion, and Fabric Notebooks offer a code-first option using PySpark, Python, Scala, R, or SQL.

A team needs to make data stored in Amazon S3 queryable from a Fabric Lakehouse without copying it into OneLake. Which Fabric capability should they use?

• AFabric Mirroring
• BA Fabric pipeline
• CFabric Eventstream
• DA OneLake shortcut✓

Answer: A OneLake shortcut is a live reference to external storage such as Amazon S3, making the data appear inside a Lakehouse without any copying or movement. Fabric Mirroring instead continuously replicates an external database such as Azure SQL Database into OneLake, which does involve copying data. A Fabric pipeline moves and transforms data as part of an ETL or ELT process, and Fabric Eventstream is built for real-time streaming sources rather than file storage like S3.

Which Microsoft Fabric component continuously replicates an external database, such as Azure SQL Database, into OneLake in near-real-time without requiring any pipeline authoring?

• AFabric Mirroring✓
• BFabric Eventstream
• CFabric Data Factory
• DA OneLake shortcut

Answer: Fabric Mirroring is configured once against a source connection and then handles change tracking and replication automatically, landing data in Delta Lake format. Fabric Eventstream instead ingests real-time streaming events from sources such as Event Hubs or Kafka. A OneLake shortcut references external data without copying it at all, and Fabric Data Factory requires building and scheduling a pipeline.

Which Fabric Data Factory component orchestrates multi-step data movement by chaining activities that run in sequence or in parallel?

• AFabric Notebooks
• BDataflows Gen2
• COneLake shortcuts
• DPipelines✓

Answer: Pipelines consist of activities that run in sequence or in parallel, using linked services to connect to sources and destinations for a multi-step workflow. Dataflows Gen2 instead provides a low-code, visual way to build transformation logic with Power Query. OneLake shortcuts reference external storage rather than orchestrating movement, and Fabric Notebooks provide a code-first alternative powered by Apache Spark.

A team is connecting to on-premises data sources in a hybrid environment and loading the results into Azure SQL Database, outside of a Microsoft Fabric workspace. Which service should they use?

• AFabric Data Factory
• BLakeflow Declarative Pipelines
• CDatabricks Notebooks
• DAzure Data Factory✓

Answer: Azure Data Factory is the standalone Azure service for building data integration pipelines outside Fabric, including connecting to on-premises sources in a hybrid environment and loading data into destinations such as Azure SQL Database. Fabric Data Factory provides the same pipeline model but operates inside a Fabric workspace rather than as a standalone service. Lakeflow Declarative Pipelines and Databricks Notebooks are Azure Databricks ingestion tools, not standalone Azure integration services.

Which Azure Databricks ingestion approach lets a developer define what the output tables should contain, while Databricks automatically handles execution ordering, dependency tracking, and incremental processing?

• ADatabricks Notebooks
• BUnity Catalog
• CFabric Mirroring
• DLakeflow Spark Declarative Pipelines✓

Answer: Lakeflow Spark Declarative Pipelines is a declarative framework in which you define what the output tables should contain, and Databricks handles execution ordering, dependency tracking, and incremental processing automatically. Databricks Notebooks instead suit ad-hoc or exploratory ingestion work where code controls each step directly. Unity Catalog is a governance layer for access control and lineage, not an ingestion mechanism, and Fabric Mirroring is a Microsoft Fabric replication feature rather than an Azure Databricks tool.