News for hackers. Stuff that matters.

A comment on Using the linear distance operator in Postgres 12 to find the closest match. Parent comment

Re:Using the linear distance operator in Postgres 12 to find... by changelink on Thursday November 07, 2019 @10:50AM (#21474220)
Postgres has a extension called cube (https://www.postgresql.org/docs/9.6/cube.html) that can be used for up to 150 dimensions (which is a compile-time limit, you can have more if you compile Postgres yourself).

It's a pretty cool extension that does distance between points, intersections between n-dimensional cubes (hence the name), different distance metrics etc.

It'd be perfect for storing and searching through large amounts of n-dimensional embeddings, I'm guessing it's used for that already.

[ Parent | on HN ]
Re:Using the linear distance operator in Postgres 12 to find... by kaivi on Thursday November 07, 2019 @03:06PM (#21476806)
In my experience, the cube extension is unusable for >10M x 128D vectors without PCA. I'm using Faiss now with ~500M vectors, and it works great!
[ Parent | on HN ]
Re:Using the linear distance operator in Postgres 12 to find... by phelmig on Thursday November 07, 2019 @10:45PM (#21479810)
With how many dimensions are you using Faiss with 100m+ vectors? I’m currently looking a solution to handle 1024 dimensions for ~100m items.
[ Parent | on HN ]
Re:Using the linear distance operator in Postgres 12 to find... by kaivi on Thursday November 07, 2019 @11:32PM (#21479990)
On one index I'm using OPQ16_64,IVF262144_HNSW32,PQ16 with 128 dimensions initially.

1024 dimensions is a lot! Could you elaborate on what application requires that many? If it's a DNN layer output, your data must be sparse, so dimensionality reduction won't affect your recall if tuned properly.

[ Parent | on HN ]
Re:Using the linear distance operator in Postgres 12 to find... by phelmig on Sunday November 10, 2019 @10:46AM (#21498408)
It's actually a DNN layer output. I haven't considered dimensionality reduction, yet. Thanks for pointing my there, I'll look into it. Probably thats the better way to go.

Thanks a lot for your reply!

[ Parent | on HN ]
Re:Using the linear distance operator in Postgres 12 to find... by orf on Thursday November 07, 2019 @01:20PM (#21475758)
Fun fact, there’s a bug in the implementation where you can create cubes with much higher dimensions because one constructor doesn’t do the check.

Still wouldn’t recommend it though.

[ Parent | on HN ]