DP-900 Describe an analytics workload on Azure Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Azure Analysis Services

  Why it's wrong here

  Azure Analysis Services is an OLAP engine that composes tabular semantic models from already-prepared data, typically loading from Azure Synapse Analytics, SQL Database, or Power BI datasets. It provides enterprise-grade in-memory compression and DAX/MDX querying for performance-critical reporting, but it does not include pipeline activities for extracting, cleansing, or transforming raw data across sources. In an ETL workflow, data must be transformed upstream before Analysis Services can even consume it.

• ✗

  Azure Databricks

  Why it's wrong here

  Azure Databricks is an Apache Spark-based unified analytics platform designed for data engineering, machine learning, and interactive big-data analytics. While it can certainly perform transformations, it does so through code-first notebooks, Spark jobs, or Python/Scala libraries, and it lacks the code-free, visually designed pipeline orchestration and mapping data flows that Azure Data Factory offers for low-code ETL. Databricks also requires you to manage clusters and runtime choices, whereas Data Factory abstracts much of the infrastructure away for batch data movement and transformation.

• ✓

  Azure Data Factory

  Why this is correct

  Azure Data Factory is the correct option because it is Azure's native, cloud-scale ETL and data integration service. It provides a visual, code-free experience through pipeline orchestration and Mapping Data Flows, which let you design data-cleaning and transformation logic such as joins, aggregations, and custom expressions without writing code. Data Factory also supports building reusable pipelines that handle multiple data sources, schedule or trigger them, and call other services like Azure Databricks or SQL Server Stored Procedures when needed.

• ✗

  Azure Stream Analytics

  Why it's wrong here

  Azure Stream Analytics is built exclusively for real-time event processing, ingesting live data from sources like Azure Event Hubs, IoT Hub, or Blob storage with timestamps. It uses a SQL-like language with temporal windowing functions to detect patterns, calculate aggregations, and smooth streaming data, but it is not intended for scheduled batch transformation or multi-source data cleansing scoped to a point-in-time ETL job. Stream Analytics is the right tool for continuous monitoring or alerting, not for the kind of on-demand, reproducible pipeline that Azure Data Factory provisions.

Go deeper

Related to this question