CockroachDB For AI/ML: Vector Indexing

Описание: In CockroachDB 24.2 ,we released pgvector support for the vector datatype, which was a good way for customers to begin experimenting with advanced LLM applications. When we introduced pgvector support, we did so for vectors with thousands of dimensions. Now customers are moving their testing along to where performance is important.

To support performant search and discovery we are now introducing vector indexing into Preview based on a protocol we’re calling Cockroach-SPANN (or C-SPANN). C-SPANN is based on Microsoft’s SPANN and SPFRESH research papers which, along with RaBitQ, deliver an indexing method that is small, fully distributed, and easy to update without degrading index quality. We’re very excited about this release as it directly addresses scalability concerns of vector applications for customers with support for indexing billions of vectors.

To show you how this works, here's a short video from our Technical Evangelist Rob Reid. Join us as he puts Vector Indexes to the test to build a demo of a trading application where he detects anomalies.

CockroachDB For AI/ML: Vector Indexing
00:00 New AI features in CockroachDB
00:29 Enabling vector indexes in CockroachDB / Configuring demo app
01:12 Creating a trigger and function pair (BEFORE INSERT trigger)
01:54 Inserting test data
02:25 Writing a function that finds anomalies
02:59 Testing this application
03:25 The possibilities of CockroachDB for AI/ML