The Microsoft Certified: Azure Data Fundamentals (DP-900) validates that you understand the basic language of data work on Microsoft Azure: relational data, non-relational data, analytics workloads, and the Azure services that support them. It is the right starting point if you want a clean mental model of modern data platforms before specializing in engineering, analytics, databases, or AI.
DP-900 is not a deep implementation exam. Microsoft is testing whether you can identify the correct data concept and map it to the right Azure service family. That means your preparation should emphasize distinctions such as transactional vs analytical workloads, relational vs non-relational data, data lakes vs data warehouses, and Azure SQL Database vs Azure Cosmos DB vs Azure Storage.
That fundamentals layer matters. If you later move toward data engineering, analytics, AI, or solution design, the concepts in DP-900 make later Microsoft exams much easier to reason about.
Exam At a Glance
| Attribute | Value |
|---|---|
| Certification | Microsoft Certified: Azure Data Fundamentals |
| Exam code | DP-900 |
| Level | Fundamentals |
| Duration | 45 minutes |
| Cost | $99 USD |
| Prerequisites | No formal prerequisite; Microsoft positions it for learners building foundational data literacy |
| Target candidate | Beginners exploring data roles, analysts, developers, students, and technical professionals who need Azure data vocabulary |
| Primary focus | Core data concepts, relational and non-relational data on Azure, and analytics fundamentals |
- Official certification page: Microsoft Certified: Azure Data Fundamentals
- Official exam page: Exam DP-900: Microsoft Azure Data Fundamentals
- Official study guide: DP-900 study guide
- Official learning paths: Explore core data concepts, Explore relational data in Azure, Explore non-relational data in Azure, Explore data analytics in Azure
Official Assessed Areas
- Describe core data concepts
- Identify considerations for relational data on Azure
- Describe considerations for working with non-relational data on Azure
- Describe an analytics workload on Azure
Like AZ-900, Microsoft's public DP-900 page emphasizes the active skill areas rather than publishing a detailed weighting table. Treat each section as exam-relevant and prepare to switch quickly between concept questions and service-selection questions.
1. Describe Core Data Concepts
This section defines the language the rest of the exam depends on. If you are shaky on data formats, workloads, and roles, the service questions later in the exam will feel harder than they need to.
- Structured, semi-structured, and unstructured data - Know what each data form looks like and why different storage systems handle them differently. Official resources: Core data concepts learning path, Explore core data concepts.
- Files, databases, and common data stores - DP-900 expects you to recognize when data belongs in files, tables, document stores, or analytical platforms. Official resource: Explore core data concepts.
- Transactional versus analytical workloads - This distinction shows up throughout the exam. Transactional systems optimize for fast day-to-day operations, while analytical systems optimize for insight and aggregation across large datasets. Official resource: Explore core data concepts.
- Roles and responsibilities in data work - Understand the basic difference between data analysts, data engineers, database administrators, developers, and architects. Microsoft uses this to frame why different services exist. Official resource: Explore data roles and services.
- Concept clarity matters more than implementation detail - This domain is about classification: identifying what kind of data or workload you are looking at before choosing a tool. Official resources: Core data concepts module, Roles and services module.
Exam tip: If a question sounds abstract, do not rush to match it to an Azure product. First decide what type of data or workload is being described. The correct Azure service choice usually becomes obvious after that.
2. Identify Considerations for Relational Data on Azure
This section covers classic relational database concepts and the Azure services used to host relational workloads.
- Relational database fundamentals - Review tables, rows, columns, keys, relationships, normalization, and database objects. Microsoft wants you to recognize what makes data relational before it asks where to run it. Official resources: Relational data learning path, Explore fundamental relational data concepts.
- SQL as the query language for relational systems - You do not need advanced query-writing skill, but you should recognize what SQL is used for and why it belongs with relational systems. Official resource: Explore fundamental relational data concepts.
- Azure relational database services - Learn the role of Azure SQL Database and the broader Azure relational services family. The exam is usually testing service positioning rather than administration depth. Official resource: Explore relational database services in Azure.
- Azure SQL Database basics - Azure SQL Database is a fully managed PaaS relational database service. Study its core positioning: managed operations, high availability, scaling options, security features, and fit for modern cloud applications. Official resource: What is Azure SQL Database?.
- Know when relational data is the right fit - If the scenario needs structured schema, joins, integrity constraints, and transactional consistency, Microsoft usually wants you thinking relational first. Official resources: Relational concepts module, Azure SQL Database overview.
Exam tip: When you see relationships, normalization, SQL querying, or transactional business applications, rule in relational systems early. The wrong answers usually drift toward file or NoSQL services that do not match the data model.
3. Describe Considerations for Working with Non-Relational Data on Azure
This section focuses on storage services and distributed data stores used when rigid relational structure is not the best fit.
- Azure Storage for files and object data - Study Blob Storage, Azure Files, Tables, and the general role of Azure Storage in handling non-relational data. Official resources: Non-relational data learning path, Explore Azure Storage for nonrelational data.
- Azure Data Lake Storage - Know that Data Lake Storage is built on Blob Storage and adds capabilities aimed at analytics workloads, including hierarchical namespace and data-lake-style organization. Official resource: Introduction to Azure Data Lake Storage.
- Azure Cosmos DB fundamentals - Azure Cosmos DB is the core Microsoft NoSQL exam topic. Understand why it fits globally distributed, highly scalable, low-latency applications and why it is positioned differently from relational databases. Official resources: Explore fundamentals of Azure Cosmos DB, Azure Cosmos DB overview.
- Choose non-relational services by workload shape - Microsoft commonly tests whether a workload needs flexible schema, massive scale, file storage, or globally distributed operational data. Official resources: Azure Storage module, Azure Cosmos DB overview.
- Separate operational NoSQL from analytics platforms - Cosmos DB is for operational data at scale, while analytical workloads typically move toward data lakes, warehouses, or analytical engines. Official resources: Azure Data Lake Storage introduction, Cosmos DB overview.
Exam tip: For non-relational questions, ask whether the workload is primarily about files, flexible-schema application data, or large-scale analytics staging. Those three patterns often separate Azure Storage, Azure Cosmos DB, and data-lake answers.
4. Describe an Analytics Workload on Azure
This section tests whether you understand how modern analytics solutions are assembled from ingestion, storage, transformation, warehousing, and visualization components.
- Data warehousing architecture - Study the basic shape of an analytics platform: ingest data, store it appropriately, transform it, model it, and expose it for reporting or analysis. Official resources: Data analytics learning path, Explore fundamentals of large-scale analytics.
- Ingestion pipelines and analytical stores - Know that analytics is not just storage. Microsoft expects you to recognize pipelines, data movement, and analytical stores as part of the broader solution. Official resource: Explore fundamentals of large-scale analytics.
- Azure Synapse Analytics positioning - Azure Synapse Analytics brings together SQL, Spark, data integration, and lake-centric analytics patterns. DP-900 usually tests its role in the analytics stack, not low-level implementation details. Official resource: What is Azure Synapse Analytics?.
- Real-time analytics and visualization basics - Microsoft's current learning path includes stream processing and visualization concepts, so be comfortable with the idea that analytics can be batch, near real-time, or dashboard-driven. Official resources: Explore fundamentals of real-time analytics, Explore fundamentals of data visualization.
- Current Microsoft Learn paths span Azure plus adjacent analytics tooling - The official training now includes broader analytics concepts and services, but the exam still centers on the fundamentals of analytical workloads and Azure data service selection. Official resources: Data analytics learning path, Azure Synapse Analytics overview.
Exam tip: If the question is about insight generation across large volumes of data, think analytics stack: ingestion, storage, transformation, warehousing, and reporting. Do not confuse that with the operational database you use to run the day-to-day application.
Recommended 4-Week Study Plan
| Week | Focus | Primary resources |
|---|---|---|
| 1 | Core data concepts, data formats, workloads, and roles | Core data concepts learning path, core data concepts module, roles and services module |
| 2 | Relational concepts, SQL basics, Azure relational services | Relational learning path, relational concepts module, Azure relational services module, Azure SQL Database overview |
| 3 | Non-relational services, Blob Storage, Data Lake Storage, Cosmos DB | Non-relational learning path, Azure Storage module, Azure Cosmos DB module, Data Lake Storage intro, Cosmos DB overview |
| 4 | Analytics workloads, warehousing, pipelines, Synapse, real-time and visualization review | Analytics learning path, modern data warehouse module, Synapse overview, stream processing and visualization modules, practice questions |
Last-Mile Exam Strategy
- Study the exam as a classification exercise. Most DP-900 questions become easier once you identify whether the workload is transactional, non-relational operational, or analytical.
- Memorize the core service-positioning contrasts: Azure SQL Database for managed relational workloads, Azure Cosmos DB for highly scalable NoSQL workloads, and Azure Storage or Data Lake Storage for file and analytical data scenarios.
- Use the official Microsoft Learn module names as your revision checklist. They map closely to the public exam outline.
- Do not spend your time on advanced administration. DP-900 rewards service recognition and data-platform fundamentals more than configuration depth.
- Review analytics terms deliberately. Many candidates are comfortable with databases but weaker on warehousing, ingestion, and visualization concepts.
If you want exam-style reinforcement after the official docs, use our DP-900 practice questions. If you want the broader Azure platform foundation first, pair this with our Azure Fundamentals study guide.
The fastest route to passing DP-900 is to study it as a map of data workload types. Learn how Microsoft describes relational, non-relational, and analytical systems, then connect each pattern to the Azure service family that fits it best. That is the core reasoning the exam is testing.