Vector support in SQL Server opens the door to modern semantic search scenarios—but embeddings alone are rarely enough. In this session, you’ll learn how to store, index, and query vector data in SQL Server, with a strong focus on high‑performance vector search using DiskANN.

We’ll go beyond theory and look at how DiskANN enables fast and scalable approximate nearest‑neighbor search directly inside SQL Server, even on large datasets. From there, we’ll combine vectors with relational data, metadata filtering, and full‑text search to build richer and more precise search experiences on your own data.

Through practical demos, you’ll see how to model data that mixes vectors and traditional columns, choose the right indexing strategy, and design queries that blend semantic similarity with exact filters and text search—without moving data out of the database.

By the end of the session, you’ll have a clear mental model of when and how to use DiskANN, how it fits into real‑world SQL Server workloads, and how to implement better search solutions using the data you already have.

You will learn:

• Understand when and why to use DiskANN in SQL Server,
• How to store, index, and query vector data alongside relational and text data
• Apply vector search to real‑world scenarios using your own data