Microsoft Fabric: Unified Analytics, Data Engineering & Real-Time Intelligence

Introduction to Microsoft Fabric

• Evolution of Microsoft data platforms
• Unified SaaS analytics vision
• Core workloads in Microsoft Fabric
• How Fabric unifies engineering, BI, and AI
• Capacity planning and pricing models
• OneLake architecture and storage abstraction
• Workspace hierarchy and governance principles
• Role-based access control (RBAC) overview
• Architecture discussion:
  • SaaS analytics vs traditional PaaS architectures
  • Multi-workload consolidation strategies
• Hands-on:
  • Explore Fabric workspace
  • Configure capacity and workspace settings

Lakehouse Architecture in Fabric

• Understanding the Lakehouse paradigm
• Delta Lake fundamentals within Fabric
• Unified experience across Spark, SQL, and Power BI
• Creating and configuring a Lakehouse
• Schema enforcement and table management
• Managing metadata and versioning
• Querying Delta tables from multiple workloads
• Hands-on:
  • Create Lakehouse
  • Ingest structured and semi-structured data
  • Query Delta tables via SQL endpoint

Dataflows Gen2 & OneLake File Management

• Introduction to Dataflow Gen2
• Low-code ingestion and transformation
• Data preparation patterns
• Managing OneLake files and folders
• Organizing enterprise data assets
• Data lineage fundamentals
• Hands-on:
  • Build a Dataflow Gen2 ingestion process
  • Explore OneLake File Explorer

SQL Analytics & External Data Integration

• SQL Analytics Endpoint architecture
• Query optimization fundamentals
• Managing external shortcuts
• Connecting to external storage systems
• Performance considerations for SQL workloads
• Security and artifact sharing
• Architecture discussion:
  • Lakehouse vs Warehouse query patterns
  • When to use shortcuts vs ingestion
• Hands-on:
  • Create SQL queries and performance analysis
  • Configure shortcut to external data source

Semantic Models & Power BI Integration

• Designing semantic models
• Relationships and star schema design
• Writing measures using DAX
• Connecting Power BI Desktop to Fabric
• Auto-generated reports and dashboards
• Managing lifecycle via Fabric Apps
• Deployment pipelines overview
• Hands-on:
  • Build semantic model
  • Create DAX measures
  • Publish report to workspace

Data Pipelines & Orchestration

• Overview of Data Pipelines
• Integration with Azure Data Factory concepts
• Copy Data activities and connectors
• Integrating Dataflows into pipelines
• Parameterization and configuration
• Scheduling and monitoring executions
• Logging and retry strategies
• Hands-on:
  • Build multi-step pipeline
  • Configure scheduling and monitoring

Data Warehousing in Fabric

• Creating and configuring a Fabric Warehouse
• T-SQL operations (INSERT, UPDATE, DELETE, COPY INTO)
• Schema evolution and computed columns
• Performance optimization strategies
• Referencing SQL endpoints from Lakehouses
• Cloning and versioning tables
• Reporting directly from Warehouse
• Architecture discussion:
  • Lakehouse vs Warehouse trade-offs
  • Workload separation strategies
• Hands-on:
  • Create Warehouse
  • Perform DML operations
  • Build report from Warehouse data

Apache Spark in Fabric

• Spark architecture fundamentals
• Working with notebooks
• Connecting Spark to Lakehouse
• DataFrame transformations
• Aggregations, joins, and window functions
• Writing SparkSQL queries
• Temporary views and persistent tables
• Python and SQL interoperability
• Scheduling Spark jobs
• Hands-on:
  • Transform dataset using PySpark
  • Persist results as Delta tables

Real-Time Analytics with KQL

• Overview of Real-Time Intelligence workload
• Creating KQL databases
• KQL fundamentals (filtering, summarization, joins)
• Ingestion techniques for streaming data
• Building Eventstreams
• Routing data to Lakehouse and KQL
• Integrating KQL with Power BI
• Architecture discussion:
  • Streaming vs micro-batch processing
  • Event-driven analytics design
• Hands-on:
  • Create Eventstream
  • Query streaming data using KQL

Data Science Workflows in Fabric

• Data Science workload overview
• Working with notebooks and Pandas
• Data Wrangler for transformation
• Exploratory Data Analysis (EDA)
  • Feature engineering strategies
  • Experiment tracking
  • Model training and evaluation (ROC, Precision-Recall)
  • Saving and deploying models
• Hands-on:
  • Train and evaluate ML model
  • Track experiment metrics

Data Activator & Intelligent Automation

• Introduction to Data Activator
• Understanding Reflex triggers
• Creating triggers from Power BI datasets
• Configuring alerts using Eventstream data
• Automated actions and notifications
• Governance and monitoring of automated workflows
• Hands-on:
  • Create Reflex rule
  • Trigger automated alert from real-time dataset

Capstone: End-to-End Unified Fabric Implementation

Participants will:

• Ingest batch and streaming data
• Implement Bronze–Silver–Gold Lakehouse layers
• Build Warehouse reporting layer
• Create Power BI dashboard
• Implement real-time Eventstream analytics
• Configure automated Data Activator trigger
• Outcome:
  • A complete unified analytics solution built entirely within Microsoft Fabric.