by otoolep on 8/15/24, 5:27 PM with 0 comments
Recently I released support for SQL Extensions[2], making it possible to load extensions into rqlite at launch time. This has opened up a bunch of new possibilities for the database. I have now taken it one step further, and baked a bunch of SQLite extensions into the rqlite Docker images[3].
All of this means that rqlite can be deployed as highly-available Vector Search database in a matter of seconds, thanks to the convenience of Docker and loading sqlite-vec[4] -- but all in a lightweight, easy-to-use manner:
Launch an rqlite node with vector search enabled:
docker run -e SQLITE_EXTENSIONS='sqlite-vec' -p4001:4001 rqlite/rqlite
Set up vector search tables using the rqlite shell:
127.0.0.1:4001> CREATE VIRTUAL TABLE vec_examples USING vec0(sample_embedding float[8])
This feels like a promising new product area for rqlite -- super easy to use, network accessible Vector Search, all in a highly available package if you want that. I'm looking for any and all feedback, as this is a very new area for rqlite (SQLite extensions, and new types of search).
[2] https://www.philipotoole.com/rqlite-8-27-loadable-sqlite-ext...
[3] https://www.philipotoole.com/rqlite-828-enhanced-docker-supp...