DP-900 Study Roadmap

Start with structured, semi-structured, and unstructured data. Practice identifying each from short scenarios because the data shape often determines whether relational tables, JSON documents, object storage, or analytics storage makes sense.

Next, compare operational systems with analytical systems. Transactional workloads update records accurately and quickly. Analytical workloads summarize, aggregate, and explore data. This distinction appears across relational database, warehouse, lake, and BI questions.

Review tables, columns, rows, keys, relationships, joins, normalization, constraints, views, indexes, and basic SQL terms. DP-900 does not require advanced SQL tuning, but it does expect you to recognize relational design and query concepts.

Map relational concepts to Azure SQL Database and other managed relational offerings. Focus on when a managed relational database fits the scenario and how relational services differ from Cosmos DB, storage accounts, or analytics platforms.

Study document, key-value, graph, and file/object storage patterns. Match Cosmos DB to globally distributed NoSQL scenarios and Azure Storage/Data Lake Storage to file and object scenarios. Avoid memorizing product names without the model behind them.

Review data warehouses, data lakes, lakehouses, ETL, ELT, batch processing, streaming, and the difference between schema-on-write and schema-on-read. These topics explain how raw data becomes curated data for analysis.

Learn what Power BI reports, dashboards, semantic models, and visualizations are for. Understand Microsoft Fabric as a broader analytics platform at a recognition level, especially where it connects integration, storage, engineering, warehousing, real-time analytics, and BI.

Use mixed questions only after the individual categories feel familiar. Classify every miss by concept: relational design, NoSQL model, Azure SQL, Cosmos DB, analytics workload, batch versus streaming, or visualization. Then revisit the weakest category before another mixed set.