Godis Hivemind

Godis Hivemind is a decentralized AI compute platform where users can contribute their computing resources to collaboratively train AI models.

Project Overview

This platform enables a peer-to-peer network for sharing CPU and GPU resources. It features a smart churn prevention system to maintain community engagement and ensure the stability of the network.

Architecture

The system is built on a microservices architecture using Go, with the following core components:

• Coordinator Service: Manages task scheduling, node registration, and network health.
• Node Service: A daemon that runs on contributor machines to execute tasks and monitor resources.
• CLI Tool: A command-line interface for users to interact with the platform.

Getting Started

Prerequisites

• Go (version 1.18 or later)
• Docker and Docker Compose
• PostgreSQL
• Redis

Setup

1. Clone the repository:

   git clone https://github.com/nothr/godis.git
   cd godis/hivemind

2. Set up the environment:

   cp .env.example .env

   Update the .env file with your local configuration details.

3. Run the setup script:

   ./scripts/setup.sh

4. Build the services:

   make build

Running the Services

• Start the coordinator:

  make run-coordinator

• Start a worker node:

  make run-node

Usage

Use the CLI to interact with the Godis network:

• Register a node:

  ./build/cli register-node --address <your-node-ip>

• Submit a job:

  ./build/cli submit-job --model-url <url-to-model>

• Check network status:

  ./build/cli status

Development

• Run tests:

  make test

• Clean build artifacts:

  make clean