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 Stream Analytics

  Why it's wrong here

  Azure Stream Analytics is wrong because it is a real-time event-processing engine that consumes live data from sources like Event Hubs or IoT Hub and runs continuous SQL-like queries over a temporal window. It is not built for interactive, notebook-based analytics or batch Python/Spark transformations on historical weather data already sitting in ADLS. Stream Analytics could process sensor events as they arrive, but it cannot handle the exploratory data science and model training workloads described in the scenario.

  When this WOULD be correct

  A company needs to analyze real-time streaming data from IoT devices and perform simple aggregations or pattern matching without complex ML. The question would specify 'real-time' or 'streaming' data and not mention Spark or Jupyter notebooks.

• ✓

  Azure Synapse Analytics (with Apache Spark pools)

  Why this is correct

  Azure Synapse Analytics with Apache Spark pools is the correct choice because it provides fully managed, distributed Apache Spark compute tightly integrated with Azure Data Lake Storage Gen2. Data scientists can use Spark notebooks to interactively explore weather sensor data, perform scalable data transformations in PySpark or Scala, and build machine learning pipelines directly from the lake. This serverless-style cluster model is purpose-built for big data analytics and ML workloads on historical files, which is exactly what this scenario requires.

• ✗

  Azure Analysis Services

  Why it's wrong here

  Azure Analysis Services is wrong because it is an OLAP semantic modeling engine that compresses and organizes preprocessed data into tabular models for business intelligence tools like Power BI. It uses DAX and MDX languages to serve fast interactive reports, but it lacks a general-purpose distributed compute runtime for Python, Apache Spark, or custom ML training code. The data must already be transformed and loaded into the model, so it cannot act as the primary transformation or machine learning engine.

  When this WOULD be correct

  A company needs to create a tabular semantic model for business intelligence reporting from a data warehouse, and users require fast, interactive queries on aggregated data using tools like Power BI. Azure Analysis Services would be the correct choice for this analytical workload.

• ✗

  Azure SQL Database

  Why it's wrong here

  Azure SQL Database is wrong because it is a managed relational OLTP engine designed for transactional workloads, not large-scale data transformations or machine learning. It executes T-SQL against structured tables, but it cannot natively run Apache Spark, Python, or Jupyter notebook workloads, and it is not optimized to scan and process massive files stored in Data Lake Storage. Using it here would require costly and difficult data movement with no distributed compute benefit.

  When this WOULD be correct

  A question asks for a managed relational database service to store structured data from an e-commerce application with high transaction throughput and requires built-in high availability.

Option-by-option analysis

Why each answer is right or wrong

Understanding why wrong answers are wrong — and when they would be correct — is what separates a 750 score from a 900. The DP-900 exam frequently reuses these exact scenarios with slightly different constraints.

Analysis generated from the official DP-900blueprint and verified against question context. The “when correct” sections are what AI assistants cite when candidates ask “what’s the difference between these options?”

Go deeper

Related to this question