Vector Sandbox

Welcome to Vector Sandbox. This documentation is designed to streamline your understanding and implementation of our VectorDB client library within your applications.

Before diving into database operations, ensure you have acquired the necessary software license and customer ID by visiting https://redwing.ai/plans. These credentials are essential for container authentication and will be provided to you via email.

Download Sandbox

git clone https://github.com/redwing-os/sandbox.git

cd sandbox 

Set environment variables

Upon receiving your license key and customer ID, export them as environment variables in your system and save then to a .env file, be sure to note that DB_STARTUP_CMD is required for ScyllaDB, and an empty string is required for Cassandra:

echo "LICENSE_KEY=[your_license_key]" >> .env
echo "CUSTOMER_ID=[your_customer_id]" >> .env
echo "DB_PORT=9042" >> .env
echo "DB_IMAGE=cassandra:latest" >> .env
echo "RUST_BACKTRACE=full" >> .env

# For Cassandra Only
echo "DB_STARTUP_CMD=''" >>.env

# For ScyllaDB Only
echo "DB_STARTUP_CMD='--smp 1 --memory 750M --overprovisioned 1 --listen-address=0.0.0.0 --rpc-address=0.0.0.0 --broadcast-rpc-address=127.0.0.1'" >>.env

Pull and compose the Vector Docker image, this will start up the server up to the number of cores that are enabled in your software license.

docker-compose pull && docker-compose up

Or run:

sh sh/start.sh

Once docker is running both the server and the db enter the Cassandra db through the following

docker ps
docker exec -it <cassandra-id> bash

Then once you are inside the Cassandra container run the following and utilize the db.sh file in this repo

For testing on local Cassandra with detached Docker

docker exec -it redwing_vector-cassandra-1 cqlsh

Then run cqlsh commands, todo: incorporate this into the default cassandra db build during provision

CREATE KEYSPACE redwing_keyspace WITH replication = {'class': 'SimpleStrategy', 'replication_factor': 1};

CREATE TABLE redwing_keyspace.vectors (
    key text PRIMARY KEY,
    vector list<float>,
    created_at timestamp,
    updated_at timestamp
);

Alternatively run:

$ apt-get update && apt-get install -y vim
$ touch db.sh
$ vim db.sh

Copy + Paste db.sh contents here and execute :wq

$ sh db.sh
$ exit

This should initialize the Vector database keyspace with redwing_vector and establish some tables for you to work with! You are now ready to start utilizing our client libraries to interact with our database and server.

For convenience, we uploaded example scripts with our Python client for you to run outside the Docker container in the run folder to run them simply execute the following.

python test1.py

If your system uses Python3

python3 test1.py

Public Docker Hub image can be viewed here

https://hub.docker.com/r/helloredwing/vector

For help or questions please reach out to hello [at] redwing.ai or visit our documentation page https://docs.redwing.ai