DP-900: Microsoft Azure Data Fundamentals

Build the foundation every Azure data role begins with — how to describe the shape of data, name the workloads that consume it, and match each Azure data service to the workload it was built for. This lesson introduces structured, semi-structured, and unstructured data; the file formats and data stores you'll meet across the DP-900 exam; the difference between transactional and analytical work; and the day-to-day responsibilities of the database administrator, data engineer, and data analyst roles. By the end of this lesson you will be able to describe the features of each data shape, identify the right Azure datastore for a given scenario, compare OLTP and OLAP workloads, and name the responsibilities that distinguish the three data roles Microsoft measures on DP-900 exam objective 1.

Ground your relational-database knowledge in the specific Azure services Microsoft measures on the DP-900 exam. This lesson takes you through the anatomy of relational data — tables, keys, and relationships — the normalization discipline that shapes production schemas, the four SQL statement categories (DDL, DML, DCL, TCL), the database objects that make schemas maintainable, the three-service Azure SQL family (SQL Database, SQL Managed Instance, SQL Server on Azure VM), and Azure's managed services for open-source relational engines (PostgreSQL and MySQL Flexible Server). By the end of this lesson you will be able to identify features of relational data, describe why normalization is used, name the SQL statement categories with examples, identify common database objects, match each Azure SQL family member to its intended scenario, and choose the right Azure managed service for an open-source relational workload.

Turn "everything that isn't a SQL table" from a fuzzy category into three concrete Azure service families you can name, size, and pick for a scenario. This lesson walks you through the Azure Storage services (Blob, Files, Table, and — for context — Queue), the blob access-tier ladder from Hot down to Archive, Data Lake Storage Gen2 as a hierarchical-namespace superset of Blob, Azure Cosmos DB's six APIs (NoSQL, MongoDB, PostgreSQL, Cassandra, Table, Gremlin) with the workload each is built for, and the partition-key mental model that makes Cosmos scale. By the end of this lesson you will be able to identify Azure Blob / Files / Table capabilities, name and place each blob access tier, distinguish Data Lake Storage Gen2 from vanilla Blob, list all six Cosmos DB APIs and pick the right one for a workload, and reason about a partition-key choice — the exact capability the DP-900 exam validates for the non-relational domain.

Learn how modern data analytics platforms turn raw operational data into curated, decision-ready insights. You'll cover ingestion, batch and streaming processing, medallion (bronze/silver/gold) organization, and the two flagship Microsoft cloud analytics platforms — Azure Databricks and Microsoft Fabric — including Fabric's Real-Time Intelligence workload for event data.

By the end of this lesson you will be able to describe the common elements of a large-scale analytics workload, compare batch and streaming approaches, and explain when to choose Azure Databricks versus Microsoft Fabric.

Learn how Microsoft Power BI turns curated analytical data into decision-ready visuals. You'll cover the Power BI product family (Desktop, Service, Mobile, Report Server, and Power BI in Fabric), semantic-model concepts (star schemas, relationships, DAX measures), and how to pick the right chart type for the business question you're trying to answer.

By the end of this lesson you will be able to describe the capabilities of Power BI, explain how semantic models organize analytical data for reporting, and identify appropriate visualizations for common analytical questions.