odyssey

Fast, Distributed Approximate Nearest Neighborhood (k-ANN) Search Engine.

Description

Odyssey: Distributed Approximate Nearest Neighborhood Search Engine

Odyssey is a horizontally scalable and distributed approximate nearest neighborhood search engine.

Odysse's goal is to provide Google production level scale and throughput, with low enough latency to be serving real time user queries, over terabytes of structured data.

Note: This project is still in the early stages of development and is not yet in a usable state.

Want to contribute? One of the easiest ways is to tell us how you're using (or want to use) Odyssey. We learn from every discussion!

Features

• supports gRPC and HTTP protocal, so you can implement applications with your favorite programming language.
• you can set any labels for each vectors, and enable to search with the label.

Status

Odyssey is at version 0.0.1 and is in developing and not ready for production usage.

Dependency

Installation

$ go get -u github.com/numb3r3/odyssey

Authors

The main authors of Odyssey are:

• Feng Wang initialized the Odyssey project and wrote its first implementation

TODO

• [] https://github.com/gorgonia/gorgonia/blob/master/walker.go
• [] https://github.com/gorgonia/vecf32
• [] https://github.com/takebayashi/ngt-cluster

Licence

Odyssey is distributed under the MIT license.