DP-900 Complete Guide: Azure Data Fundamentals Scope, Study Resources & Pass Strategy

Microsoft Certified: Azure Data Fundamentals (DP-900) is the entry-level data exam in Azure's Fundamentals tier. Where AZ-900 covers cloud in general, DP-900 narrows in on three areas — SQL databases, NoSQL databases, and data analytics platforms — at a conceptual level, positioning it as the first certification for anyone aiming to become a data engineer, data analyst, or data scientist. This article walks through the exam domains, the new Microsoft Fabric coverage, the route to a free Virtual Training Day voucher, and a study roadmap distilled from candidates who passed, all in one place.

The biggest payoff from earning DP-900 is getting a map of Azure's data services in your head. Azure ships a long list of data services — SQL Database, SQL Managed Instance, Cosmos DB, PostgreSQL, Synapse Analytics, Fabric, Databricks, Data Factory, Power BI — and beginners are often dropped into projects without a clear sense of "which one do I pick?" DP-900 is the fastest structured way to learn that selection logic, and it is optimized for building the shared vocabulary data roles use day to day.

DP-900 Exam Specifications

DP-900's specs match AZ-900. It is a 45-minute, 40-60 question multiple-choice exam, the passing score is 700 out of 1000, and the fee is $99 USD (JPY 13,200 incl. tax in Japan). It is available in 13 languages including Japanese, and you can take it either online via Pearson VUE's OnVUE or at a testing center. Like the other Fundamentals exams, it never expires — once you pass, you never need to retake it.

Domain 1: Core Data Concepts (25-30%)

The first domain tests the foundational vocabulary you need to talk about data itself. The core topics are the three classifications — structured, semi-structured, and unstructured — their representative formats (CSV / JSON / XML / images / video), the relationship between data sources and data sinks, and the difference between OLTP and OLAP.

A staple of this domain is questions about data roles and responsibilities. A Database Administrator (DBA) runs the production database (backups, performance, security), a Data Engineer designs data pipelines (ingest, transform, store), and a Data Analyst extracts business insight (BI reports, KPI dashboards). Expect scenarios that ask which role owns a given Azure-service workload. For example, "Who is responsible for ETLing 10 TB of logs every day into a data lake?" — you need to confidently pick Data Engineer.

Domain 2: Relational Data (20-25%)

This domain covers relational databases on Azure. The core is choosing between Azure SQL Database (fully managed PaaS), Azure SQL Managed Instance (PaaS closer to SQL Server, designed for migrations), and SQL Server on Azure VM (IaaS, full control). On top of that, you will see Azure Database for PostgreSQL / MySQL / MariaDB for managed open-source databases, and Azure Database Migration Service for migrations.

This domain also tests the theoretical underpinnings of relational databases: the four ACID properties (Atomicity / Consistency / Isolation / Durability), the basics of normalization, and the roles of primary keys, foreign keys, and indexes. You are not asked to write SQL, but you need to be able to read simple SELECT, JOIN, and GROUP BY syntax. For hands-on practice, spin up a single Azure SQL Database with the AdventureWorks sample DB from the Azure portal and run one SELECT through the Query Editor — the terminology in this domain sticks much better afterward.

Domain 3: Non-Relational Data (15-20%)

The smallest domain by weight, but it concentrates Azure's NoSQL services — which beginners often find the hardest to pick up. The core is Azure Cosmos DB: knowing when to use each of the 5 API families (NoSQL API, MongoDB API, Cassandra API, Gremlin API, Table API), and recognizing the 5 consistency levels (Strong, Bounded staleness, Session, Consistent prefix, Eventual) — both come up frequently.

Consistency levels become much easier once you remember a single principle: "stronger consistency trades off against higher availability." Strong guarantees full consistency but adds latency, Eventual is the fastest but only converges eventually. Anchor on the two ends of that spectrum and remember that Session (the default — order is guaranteed within a single session) is the de facto standard in real-world use, and you will handle nearly every consistency question in this domain.

Outside Cosmos DB, this domain also covers Azure Table Storage (simple key/value) and Azure Blob Storage (object storage for documents, images, and video). The Blob storage tiers (Hot / Cool / Cold / Archive) and how to choose between them are a frequent topic — and one that overlaps with AZ-900.

Domain 4: Analytics Workloads (25-30%)

The most heavily weighted domain, covering Azure's entire analytics platform stack. The core is Azure Synapse Analytics (the analytics platform that absorbed the legacy SQL DW) and, since 2024, Microsoft Fabric (a SaaS that unifies Power BI Premium, Synapse, and Data Factory).

Microsoft Fabric is the newest and most important topic in this domain. Fabric's center of gravity is OneLake, a unified data lake — Lakehouse, Warehouse, Real-Time Intelligence, Data Factory in Fabric, and the other workloads all share the same underlying data layer. The defining idea is "run BI and ML against the same data without moving it," and it sits at the heart of Microsoft's forward data strategy. DP-900 does not test deep implementation, but you do need to be able to answer "what does this service do?" and "what is it best for?" at an entry level.

Other sub-topics in the analytics workload domain include Azure Databricks (a data + ML platform built on Apache Spark), Azure Data Factory (data pipeline construction), Power BI (BI reports and dashboards), and Stream Analytics (streaming analytics). Slot each service onto two simple axes — "batch vs. streaming" and "ETL vs. visualization" — and you will handle the scenario questions much more easily.

Can You Pass with Official Resources Alone?

Like AZ-900, DP-900 is an exam you can pass using Microsoft's official resources alone. Working through the three resources below in order is enough to get even an IT beginner into pass range within 4 weeks.

With the Virtual Training Day route, you can legitimately drop the JPY 13,200 fee to effectively zero — meaning you can pass self-study and entirely free. This domain in particular (analytics workloads, including Fabric) has been revised heavily in recent years, so sticking with official resources rather than third-party books is the safer call.

4-Week Pass Roadmap

A 4-week plan aimed at IT beginners. Week 1: work through Domain 1 (core data concepts) on Microsoft Learn — pin down the boundaries between structured / semi-structured / unstructured and the responsibilities of each data role. Week 2: study Domain 2 (relational data) and spin up a single AdventureWorks sample DB in the Azure portal to run a few SELECTs. Week 3: drill Domains 3 + 4 (non-relational + analytics) intensively, and attend Microsoft Virtual Training Day Part 1 + 2 to earn the voucher. Week 4: finishing kick — drill the official Practice Assessment until you can hit 80% or higher, and patch any weak areas with targeted Microsoft Learn modules.

What to Aim for After DP-900

DP-900 is the gateway into Azure's data certification family. Three directions are available depending on your goal. For data engineering, take DP-203 (Data Engineer Associate) for pipeline design centered on Synapse, ADF, and Databricks. If you are mostly working on greenfield projects, DP-700 (Fabric Data Engineer Associate) is also strong — it lines your career up with Microsoft's forward strategy. For database administration, take DP-300 (Database Administrator Associate), which focuses on operating Azure SQL Database and Managed Instance. If AI/ML is also on your radar, running AI-901 in parallel at the Fundamentals stage makes the eventual jump to DP-203 or the PMLE track much smoother.

Frequently Asked Questions

The exam information in this article is based on the official Microsoft Learn DP-900 page and the official Study Guide. This article is not an official Microsoft Corporation product and has no affiliation or sponsorship relationship. Microsoft, Azure, Microsoft Fabric, Azure SQL, Azure Cosmos DB, and Power BI are trademarks of the Microsoft group of companies. Information is based on official public materials as of May 24, 2026. Always confirm the latest details on the official page.