Browse Certification Practice Tests by Exam Family

Microsoft DP-900 Cheat Sheet: Azure Data Fundamentals

May 1, 2026

Review the Microsoft Azure Data Fundamentals (DP-900) scope, relational and non-relational data, analytics workloads, Azure data services, Power BI concepts, and service-selection traps before practicing.

On this page

DP-900 is a fundamentals exam. Use this cheat sheet to keep basic data concepts straight before moving into Azure SQL administration, Fabric data engineering, analytics engineering, or Databricks routes.

Use this with practice. Review the Azure data fundamentals checklist, then return to the DP-900 exam page for sample questions and update tracking.

Open DP-900 practice page Compare Azure routes

Exam snapshot

FieldDetail
IssuerMicrosoft
Certification laneAzure Data Fundamentals
Exam codeDP-900
Main scopeCore data concepts, relational data, non-relational data, analytics workloads, and Azure data services
IT Mastery statusSample questions available

Fundamentals map

AreaWhat to knowCommon trap
Core data conceptsStructured, semi-structured, unstructured, batch, streaming, transactional, and analytical workloadsConfusing operational transactions with analytical reporting
Relational dataTables, rows, keys, relationships, SQL, Azure SQL, and SQL Server optionsUsing relational terms for document or key-value data
Non-relational dataDocuments, key-value, graph, wide-column, object storage, and Cosmos DB patternsAssuming flexible schema means no design decisions
Analytics workloadsIngestion, transformation, warehousing, lakehouse, visualization, and reportingTreating Power BI as the data store
Service selectionAzure SQL, Cosmos DB, Storage, Fabric, Databricks, Synapse-style concepts, and Power BIChoosing a service by name recognition instead of workload fit

Must-know distinctions

DistinctionHow to decide
Transactional vs analyticalTransactional systems run business operations; analytical systems aggregate and analyze data.
Structured vs semi-structuredStructured data has fixed tables; semi-structured data has flexible fields, often JSON.
Relational vs documentRelational data uses tables and relationships; document data stores flexible records.
Data lake vs data warehouseLakes store flexible data; warehouses organize curated analytical data.
Power BI vs databasePower BI visualizes and models data; it is not the primary operational database.

High-yield checklist

  • Start by classifying the workload: transactional, analytical, batch, streaming, relational, non-relational, or visualization.
  • Use relational databases for structured data with relationships and SQL querying.
  • Use document or other non-relational stores when records vary by shape or access pattern.
  • Use analytics services when the goal is reporting, trends, or decision support.
  • Match storage and compute choices to query pattern, latency, scale, and governance needs.
  • Know that security, access, and data quality still matter in fundamentals questions.
  • Understand basic Power BI role as visualization and reporting, not raw data storage.

Common traps

  • Calling every database workload analytics.
  • Selecting a data lake when the scenario asks for structured transactional processing.
  • Selecting SQL only because the data is important.
  • Ignoring data shape when choosing a store.
  • Treating visualization as data engineering.
  • Forgetting that fundamentals questions often test vocabulary and service fit.

Practice strategy

For DP-900 misses, write the workload type first, then the best Azure service family. If you cannot explain why a workload is relational, non-relational, transactional, analytical, batch, or streaming, drill that vocabulary before moving into role-based data exams.

Revised on Monday, May 25, 2026