Edge Node Launcher

Edge Node Launcher is a desktop application designed to manage the containerized edge node of Naeural Edge Protocol. It allows you to edit environment files, check Docker availability, launch and stop Docker containers, plot data from JSON files, and perform other management tasks. T

Features

Edit .env files for Docker container configuration
Check if Docker is available and prompt for installation if not
Launch and stop Docker containers
Display plots for CPU load, memory load, GPU load, and GPU memory load
Refresh plots every 10 seconds
Copy local node address
Delete specific files and restart containers
Toggle between dark and light themes

Binary installation

Download from Releases page and run the app for your platform.

Source code install

To install and run Edge Node Launcher, follow these steps:

Clone the repository:

git clone https://github.com/NaeuralEdgeProtocol/edge_node_launcher.git
cd edge_node_launcher

Install the required dependencies:
```
pip install -r requirements.txt
```
Run the application:
```
python main.py
```

Building the Application

The repository uses GitHub Actions to build and release binaries for different platforms (Windows, Ubuntu 22.04, and Ubuntu 20.04). The build process includes:

Checking the version in ver.py and comparing it with the latest release version
Building the application using PyInstaller
Zipping the built artifacts and uploading them as release assets

Usage

Editing Environment Files

To edit the .env file used by the Docker container, click the "Edit .env File" button. The file will open in a text editor within the application. After making changes, save the file and close the editor.

Checking Docker Availability

The application checks for Docker availability at startup. If Docker is not installed, you will be prompted to install it.

Launching and Stopping Docker Containers

Use the "Launch Container" button to start the Docker container. If the container is already running, the button will show "Stop Container" to allow stopping the container.

Viewing Plots

The application displays plots for CPU load, memory load, GPU load, and GPU memory load. The plots are updated every 10 seconds with data from a JSON file.

Updating the Application

The application can check for updates, download the latest release, and replace the current executable. To check for updates, ensure you have an internet connection and the application will handle the rest.

Development

Code Structure

main.py: Entry point for the application
app_forms/frm_main.py: Main application form and logic
utils/const.py: Constants used throughout the application
utils/docker.py: Docker utility functions

Stylesheets

The application uses stylesheets for UI theming. The stylesheets are defined in utils/const.py and applied to various widgets.

Citation

If you use this software in your research, please cite the following paper:

@misc{milik2024naeuralaios,
  title={Naeural AI OS -- Decentralized ubiquitous computing MLOps execution engine}, 
  author={Beatrice Milik and Stefan Saraev and Cristian Bleotiu and Radu Lupaescu and Bogdan Hobeanu and Andrei Ionut Damian},
  year={2024},
  eprint={2306.08708},
  archivePrefix={arXiv},
  primaryClass={cs.AI},
  url={https://arxiv.org/abs/2306.08708},
}

Additionally, you can cite this repository as follows:

@misc{naeurallauncher2024,
  author = {Naeural Edge Protocol},
  title = {Edge Node Launcher},
  year = {2024},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/NaeuralEdgeProtocol/edge_node_launcher}},
}

License

This project is licensed under the MIT License - see the LICENSE file for details.