From the course: Learn Databricks GenAI
Unlock the full course today
Join today to access over 24,700 courses taught by industry experts.
Create Vector Search index - Databricks Tutorial
From the course: Learn Databricks GenAI
Create Vector Search index
- [Instructor] Now let's look at the mechanics of how to create and query a vector search index in Databricks. I've used this article as a reference, and it pretty much works well the way it's written. So I would recommend that you actually run through this in addition to watching this movie. It does take about 30 minutes for this to complete in the infrastructure I had, anyway. So plan for 30 minutes if you want to set this up and run it on your own, or you can just watch the results here. So first of all, you have to make sure that your environment meets the requirements. You have to have the Unity catalog enabled in your workspace. Serverless compute has to be enabled, and that requires some setup steps. The source table has to have change data feed enabled, and that's for the syncing. To create a vector search index, you have to have create table privileges on the catalog, and to query the index if it's owned by another user, you got to have additional privileges. So what I did is…