What is SeahorseDB?

Seahorse DB is a high-performance vector database designed for fast and efficient similarity searches in large-scale datasets. Whether you're dealing with embeddings from machine learning models or other vectorized data, Seahorse DB provides a scalable solution to store, search, and retrieve vectors with minimal latency.

SeahorseDB Setup Guide (Docker)

This guide provides instructions to run SeahorseDB using Docker.

---

Prerequisites

Make sure you have Docker and Docker Compose installed on your system.

• Install Docker
• Install Docker Compose

---

Build Docker Image

First, clone the repository and initialize submodules:

git clone https://github.com/SeahorseDB/SeahorseDB.git
cd SeahorseDB
git submodule update --init --recursive

Then build the Docker image from the db_engine directory:

docker build -t seahorse:opensource db_engine/

---

How to Run SeahorseDB

Start with Docker Compose

docker-compose up

This will start SeahorseDB on port 5555.

Stop

docker-compose down

Configuration

You can customize SeahorseDB behavior by editing seahorsedb.conf:

appendonly yes
save ''
port 5555
loglevel notice
bg-read-command-thread 16
enable-in-filter yes

The configuration file is mounted into the container via docker-compose.yaml.