I have a huge list of vectors (extracted from images, text, audio, whatever) and I want to do metric search. I want to input a query vector and retrieve the closest ones to it according to the euclidean distance. I know there are in-memory structures like the BallTree but I was wondering if there is a Postgres-like software that could manage indexes larger than memory.

  • Geospatial databases do this, but generally with a restriction to 2- or 3-dimensional space, which probably won't suffice for your purpose. – Kodiologist Dec 4 '18 at 16:28

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.