alexklibisz / elastiknn

Elasticsearch plugin for nearest neighbor search. Store vectors and run similarity search using exact and approximate algorithms.

Home Page:https://alexklibisz.github.io/elastiknn

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Whats the feature parity between AWS ES KNN feaure and elastiknn?

anannaya opened this issue · comments

This is just a question , We would like to move from AWS ES (mainly KNN) to opensource ES with KNN plugin.
Would like to know is the features are same ?

Hmm, I don't have an exact comparison handy. Dmitry Kan has a nice article from about a year ago: https://towardsdatascience.com/speeding-up-bert-search-in-elasticsearch-750f1f34f455. Elasticsearch core has also recently released an ANN feature preview: https://www.elastic.co/blog/introducing-approximate-nearest-neighbor-search-in-elasticsearch-8-0. Beyond that, I would recommend just reading the docs and taking a couple hours to setup some indexes and try some searches.

(Feel free to re-open if you have more specific questions, e.g., "is X/Y/Z supported?")