This repository contains the overview and materials for the FabCon SQL Server 2025 Pre-Conference Training. This full-day session covers the new features, architecture, and capabilities of SQL Server 2025, focusing on its AI integration, developer tools, and hybrid cloud capabilities.
The demos referenced in this training are hosted in the following Microsoft repositories:
- Primary Demo Repo:
microsoft/bobsql/tree/master/sql2025book- Look for demos marked with 🔬 in the outline.
- Supplemental Demos:
microsoft/bobsql/tree/master/demos/sqlserver2025- Look for demos marked with 🧪 in the outline.
The training is divided into several modules covering the "ground to cloud" story of SQL Server 2025.
- Introduction to SQL Server 2025 as the AI-ready enterprise database.
- Overview of the 5 pillars: AI Built-In, Security & Performance, Developer Experience, Cloud Agility, and Fabric Integration.
- Architecture deep dive and new tooling (SSMS 22, VS Code extensions).
- Vector Search: Understanding RAG patterns, embeddings, and vector stores.
- Native AI Features:
vectordata type,VECTOR_DISTANCE, andCREATE EXTERNAL MODEL. - Building scalable AI applications and "Agentic RAG" using T-SQL.
- REST API: Automatic REST endpoints for stored procedures.
- JSON Enhancements: Native JSON type and performance improvements.
- Regular Expressions: Native regex support in T-SQL.
- GitHub Copilot: AI assistance for SQL development in VS Code and SSMS.
- Mirroring: Near-real-time data replication to Fabric.
- OneLake: Accessing SQL Server data in OneLake.
- Hybrid Analytics: Combining operational data with advanced analytics in Fabric.
- Security: Microsoft Entra authentication, Ledger, and Defender for SQL.
- Performance: Intelligent Query Processing (IQP) Next Gen, Optimized Locking.
- High Availability: Always On availability group enhancements.
To run the demos associated with this training, you will typically need:
- SQL Server 2025 (Developer Edition or higher).
- SQL Server Management Studio (SSMS) 22 or Azure Data Studio.
- Visual Studio Code with the mssql extension.
- Azure Subscription (required for features involving Azure OpenAI, Arc, or Fabric).
- Python/Notebook Environment (for some AI-related demos).
Check overview.md for the detailed slide-by-slide outline.