Microsoft Fabric Explained for DP-900: Everything You Need to Know
Microsoft Fabric is the newest addition to the DP-900 exam. It's Microsoft's unified analytics platform — and it's increasingly important to understand how it relates to (and differs from) Synapse, Power BI, and other Azure data services.
DP-900 Exam Coverage
Microsoft Fabric appears in Domain 4: Describe an analytics workload on Azure (25% of exam). The exam expects you to know what Fabric is, its core components (OneLake, Lakehouse, Warehouse, Data Factory, Power BI in Fabric), and when to position it vs standalone Azure Synapse. You won't be tested on detailed Fabric administration — just conceptual understanding.
What is Microsoft Fabric?
Microsoft Fabric is an all-in-one analytics platform that brings together data engineering, data integration, data warehousing, real-time analytics, and business intelligence into a single unified SaaS service.
Before Fabric, organizations had to stitch together multiple Azure services: Azure Data Factory for pipelines, Azure Synapse Analytics for data warehousing, Azure Databricks for Spark processing, and Power BI for visualization. Each service had separate administration, separate pricing, and separate governance.
Microsoft Fabric unifies these capabilities under one roof — with shared storage (OneLake), shared governance, and a single licensing model (Fabric capacity).
Key Exam Point
Fabric is positioned as Microsoft's strategic direction for analytics. The exam recognizes it as a "unified analytics platform" — the key differentiator from standalone services like Synapse or ADF. Think of it as: Fabric is to analytics what Microsoft 365 is to productivity — a suite of capabilities under one umbrella.
OneLake: The Foundation of Microsoft Fabric
OneLake is the unified data lake that underlies all of Microsoft Fabric. It's one of the most important concepts for the DP-900 exam.
One lake for all Fabric data
Every Fabric workspace automatically uses OneLake as its storage foundation. You don't create separate storage accounts — all data stored by Fabric items lands in OneLake automatically.
Built on Azure Data Lake Storage Gen2
OneLake is physically built on ADLS Gen2 (Azure Data Lake Storage Generation 2) — the same technology as standalone Azure Data Lake. This means it inherits ADLS Gen2's performance and compatibility.
Delta format by default
OneLake stores data in Delta Lake format (open-source, column-oriented format). This makes data immediately accessible to any Fabric compute engine — SQL, Spark, or Power BI Direct Lake — without copying or transforming.
Shortcuts
OneLake shortcuts allow Fabric to reference data stored in other locations (Azure Data Lake, Amazon S3, Google Cloud Storage) without physically moving the data. This is called "data virtualization."
Organization-wide single copy
Within an organization, all Fabric workspaces share one OneLake instance. Data is logically separated by workspace and item — but stored once. This eliminates data silos and duplication.
Microsoft Fabric Workloads (Core Components)
Fabric is organized into workloads — each a set of capabilities for a different analytics persona. The DP-900 exam expects you to recognize these workloads and their purpose:
Data Engineering
Spark-based data transformation and processing. Engineers use Notebooks and Spark jobs to ingest, clean, and transform large datasets. Similar to Azure Databricks in capability.
Key item: Lakehouse — a Fabric item that combines data lake storage with SQL analytics
Data Factory
Data integration and pipeline orchestration within Fabric. Copy data between sources, schedule pipelines, and apply simple transformations. A next-generation version of Azure Data Factory built into Fabric.
Key item: Pipelines and Dataflows Gen2 for data movement and transformation
Data Warehouse
Enterprise-grade SQL data warehousing with T-SQL compatibility. Supports complex analytical queries on large datasets. Shares OneLake storage with other Fabric workloads.
Key item: Warehouse — a fully managed SQL DW with open Delta format storage
Real-Time Intelligence
Streaming data ingestion and real-time analytics. Ingest IoT data, log streams, and events. Query data in real-time using KQL (Kusto Query Language) or Eventstreams.
Key item: Eventhouse (KQL database) and Eventstream for real-time data
Power BI
Business intelligence and reporting. Power BI is fully integrated into Fabric — reports, dashboards, and semantic models are first-class Fabric items. The "Direct Lake" connection mode gives Power BI instant access to OneLake data.
Key item: Semantic models (formerly Power BI datasets), reports, and dashboards
Data Science
Machine learning model development, training, and deployment using Python notebooks and MLflow. Integrated with OneLake for data access and Spark for distributed training.
Key item: ML models and experiments stored in OneLake
Microsoft Fabric vs Azure Synapse Analytics
This is the most likely exam comparison question involving Fabric. Both provide analytics capabilities, so when does the exam expect you to choose Fabric vs Synapse?
| Aspect | Azure Synapse Analytics | Microsoft Fabric |
|---|---|---|
| Type | Azure PaaS service | SaaS platform (subscription-based) |
| Storage | Azure Data Lake Storage (separate) | OneLake (unified, built-in) |
| SQL Capability | Dedicated + serverless SQL pools | Warehouse + SQL endpoint on Lakehouse |
| Spark | Synapse Spark pools | Data Engineering Notebooks / Spark jobs |
| Power BI | External integration | Natively integrated, Direct Lake connection |
| Data Movement | Synapse Pipelines (ADF-based) | Data Factory workload + Dataflows Gen2 |
| Governance | Synapse Studio workspace | Microsoft Purview integration |
| Positioning | Established, mature, IaaS-leaning | Strategic direction, SaaS, unified experience |
| License | Pay per resource (SQL DWU, vCores) | Fabric capacity (F-SKU) |
For DP-900: When a question asks about a "unified analytics platform" that brings together data engineering, warehousing, and BI in one place → answer is Microsoft Fabric. When a question focuses on existing Azure infrastructure, SQL pools, or an established data warehouse → answer is likely Azure Synapse.
Key Microsoft Fabric Concepts to Memorize for DP-900
| Term | Definition | Exam Relevance |
|---|---|---|
| OneLake | Unified data lake storage shared across all Fabric workloads | High — core architectural concept |
| Lakehouse | Fabric item combining data lake storage with SQL analytics endpoint | High — combines lake + warehouse paradigms |
| Warehouse | Fully managed SQL data warehouse in Fabric | Medium — T-SQL analytics at scale |
| Semantic Model | Data model for Power BI reports (formerly "dataset") | Medium — renamed in Fabric era |
| Direct Lake | Power BI connection mode that reads directly from OneLake (no import) | Medium — performance benefit |
| Shortcuts | Pointers to external data without copying it into OneLake | Medium — data virtualization concept |
| Capacity | Fabric licensing unit (F-SKU) that determines compute resources | Low — mentioned in cost scenarios |
| Workspace | Container for Fabric items, analogous to a team or project workspace | Low — organizational concept |
Practice DP-900 Analytics Questions
Our DP-900 question bank covers Microsoft Fabric, Synapse, Databricks, and all analytics workload topics. Start with 40 questions free.
Start Free DP-900 Practice →Related Resources
DP-900 Study Guide 2026
Complete DP-900 exam prep — the analytics workloads domain (Domain 4) covered in full.
Relational vs Non-Relational Azure Data
Core DP-900 concept — how to identify the right Azure data storage solution.
Free DP-900 Practice Questions
25 free questions including analytics and Microsoft Fabric scenarios.
Is DP-900 Worth It in 2026?
Salary data and career impact analysis for Azure Data Fundamentals.