DP-600 - Microsoft Fabric Overview
Quick Navigation
What is Microsoft Fabric?
Microsoft Fabric is an end-to-end, unified analytics platform that combines capabilities from multiple Azure data services into a single, integrated SaaS product. It includes:
- Data Engineering (Lakehouse, Spark Notebooks, Data Pipelines)
- Data Integration (Dataflow Gen2, Data Factory pipelines)
- Data Warehousing (Fabric Data Warehouse)
- Real-Time Analytics (Eventhouse, KQL databases)
- Data Science (ML Experiments, Models)
- Business Intelligence (Power BI reports, Paginated Reports)
OneLake
OneLake is Fabric's single, unified data lake - built on Azure Data Lake Storage Gen2. Key properties:
- One OneLake per Fabric tenant (cannot have more)
- All Fabric items (Lakehouses, Warehouses, KQL databases) store data in OneLake automatically
- Hierarchical namespace:
tenant/workspace/item/ - Accessible via Azure Storage APIs, OneLake File Explorer (Windows), ADLS Gen2 compatible endpoints
- Shortcuts: Virtual links inside OneLake that reference data in other locations (other OneLake paths, ADLS Gen2, Amazon S3, Google Cloud Storage) without copying
OneLake File Explorer is a Windows application that mounts OneLake as a drive, allowing file drag-and-drop from Windows Explorer into Fabric items.
Fabric Capacity and SKUs
Fabric runs on Fabric capacity, measured in Capacity Units (CU):
| SKU Type | Name | Description |
|---|---|---|
| F-SKU | Fabric SKU (F2, F4, F8, ... F2048) | Azure-based pay-per-use Fabric capacity. Supports all Fabric workloads. Can be paused/resumed. Purchased on Azure portal. |
| P-SKU | Power BI Premium (P1, P2, P3, P4, P5) | Existing Power BI Premium capacity. Also runs Fabric workloads after Fabric is enabled. Fixed monthly cost. |
| Trial | Fabric Trial | 60-day free Fabric capacity (F64 equivalent) for evaluation. |
A Fabric capacity admin manages the capacity: assigns workspaces to capacity, controls workload settings, and monitors CU usage via the Capacity Metrics app.
Workspaces
Workspaces are containers for Fabric items. Important attributes:
- Each workspace must be assigned to a Fabric/Premium capacity to use Fabric features
- Four roles: Admin, Member, Contributor, Viewer
- Workspaces map to deployment pipeline stages (Dev / Test / Prod)
- Git integration: a workspace can be connected to an Azure DevOps or GitHub repository branch for version control of Fabric items
When Git integration is enabled, changes to supported items (semantic models, notebooks, data pipelines) are synced with the repo. This enables collaboration, pull requests, and CI/CD workflows.
Fabric Items Summary
| Item | Purpose | Storage Format |
|---|---|---|
| Lakehouse | Open format data lake with SQL endpoint and Delta tables | Delta/Parquet on OneLake |
| Data Warehouse | T-SQL fully managed warehouse with DML support | Delta/Parquet on OneLake |
| Eventhouse / KQL Database | Real-time analytics, time-series, log data with KQL | Columnar store on OneLake |
| Dataflow Gen2 | No-code/low-code ETL with Power Query UI | Output to Lakehouse/Warehouse/etc. |
| Data Pipeline | Orchestrate data movement and transformation activities | Configuration only; data flows through activities |
| Notebook | Spark-based code (Python/Scala/SQL/R) on Lakehouse data | Code; output data in Delta on Lakehouse |
| Semantic Model | Analytical data model for Power BI reports | VertiPaq in-memory / Direct Lake |
| Report | Interactive visualizations built on a semantic model | Metadata / visuals |
| Paginated Report | Pixel-perfect print/export reports (RDL) | RDL definition |
| Data Activator | No-code alerting / action trigger on real-time data streams | Stream processing rules |
Choosing the Right Fabric Storage Item
| Scenario | Best Item |
|---|---|
| Batch ETL, Delta Lake transformations, Power BI Direct Lake | Lakehouse |
| Full SQL DML (INSERT/UPDATE/DELETE/MERGE), BI star schema, T-SQL users | Data Warehouse |
| IoT telemetry, logs, time-series, KQL analytics | Eventhouse (KQL Database) |
| No-code data transformation with Power Query | Dataflow Gen2 |
| Complex orchestration (branch, loop, parallel activities) | Data Pipeline |
| Spark ML, notebook-based data science | Notebook |
Capacity Metrics App
The Microsoft Fabric Capacity Metrics app is a pre-built Power BI app that monitors CU consumption across all workloads assigned to a Fabric capacity. It shows:
- CU utilization percentage over time
- Top items consuming capacity
- Throttling events (when burst capacity is exceeded)
- Per-operation CU usage breakdown
Capacity admins use the Metrics app to identify expensive operations, optimize workload distribution, and decide whether to scale up capacity.
Data Activator
Data Activator (also called Reflex) is a Fabric item that monitors data streams or reports and triggers actions when defined conditions are met - without writing code. Use cases:
- Send an alert email when a KPI drops below a threshold
- Trigger a Power Automate flow when a sensor reading spikes
- Create a Teams notification when a report visual crosses a value
Data Activator can monitor data from Power BI report visuals, Eventstream, or KQL databases. You define objects (the entity being monitored), properties (the metric), and triggers (the condition + action).
← Take DP-600 Practice Tests | Back to Study Topics