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 SQL Database

  Why it's wrong here

  Azure SQL Database is a relational database-as-a-service offering designed for OLTP workloads with a fixed logical storage limit (100 GB for single databases, with hyperscale scaling but still not suited for petabyte-scale data lake analytics). It does not provide a built-in way to directly query Parquet or other files in Azure Data Lake Storage; instead, data must be imported via pipelines like Azure Data Factory, which is impractical for petabyte-scale data. Its query engine is optimized for row-based transactional patterns, not for large-scale analytical scans over external data files.

  When this WOULD be correct

  For a scenario where a company needs a fully managed relational database with built-in high availability, automatic backups, and requires transactional consistency for OLTP workloads, such as an e-commerce application's order processing system.

• ✓

  Azure Synapse Serverless SQL pool

  Why this is correct

  Azure Synapse Serverless SQL pool is purpose-built for on-demand T-SQL queries over data stored in Azure Data Lake Storage (ADLS) or other open formats like Parquet. It automatically scales compute resources to match the size and complexity of the query, eliminating capacity planning and enabling petabyte-scale transactional analytics without provisioning dedicated infrastructure. Because it reads data directly from the lake via a T-SQL endpoint, it provides a familiar relational interface while preserving the cost benefits of a serverless model.

• ✗

  Azure HDInsight

  Why it's wrong here

  Azure HDInsight is a managed big data analytics service that deploys clusters of virtual machines running open-source frameworks like Hadoop, Spark, Hive, or HBase. To query the transaction data, you would need to provision and manage a cluster, choose the right instance types and node counts, and handle scaling and patching — none of which are required by a serverless SQL solution. Moreover, cluster startup and teardown times add latency, and idle compute continues to incur costs, so it is not the optimal choice for ad-hoc or interactive SQL queries over petabytes in a data lake.

  When this WOULD be correct

  An exam scenario where Azure HDInsight would be correct: A company needs to run batch ETL jobs on petabytes of unstructured data using custom MapReduce or Spark code, and they require full control over the cluster configuration and libraries.

• ✗

  Azure Databricks

  Why it's wrong here

  Azure Databricks is an Apache Spark-based analytics platform that can query data lake files using Spark SQL, DataFrames, or notebook-based Python/Scala code. However, it requires the creation and management of an interactive or job cluster with configured auto-scaling policies, and the primary interface is notebooks or Spark jobs rather than a purely serverless T-SQL endpoint. Databricks is excellent for complex data engineering and machine learning workloads, but for a simple, on-demand SQL query scenario over petabyte-scale Parquet data, Synapse Serverless SQL pool offers a lower-friction, SQL-native experience.

  When this WOULD be correct

  Azure Databricks would be correct if the question required running advanced analytics, machine learning, or ETL pipelines using Spark, and the analysts needed to perform iterative data exploration or model training rather than just SQL queries. For example: 'A data science team needs to train a machine learning model on petabytes of transaction data using Python and Spark, with the ability to collaborate in notebooks.'

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