Local Carbon Footprinter is light-weight tool designed to monitor machine usage and calculate its carbon impact. This tool is currently under active development, with the intention to output the raw metrics to a local flask endpoint (Phase I), integrate with the Cloud Carbon Footprint tool as an On-Premise data source. (Phase II), and finally, to create a user-friendly desktop application (Phase III).
Phase IV: integrate as a module in the GovA11y & Equalify Ecosystems.
Please note: This tool is currently in Phase I.
Make sure you've got docker...
TODO Dockerfile... TODO Refactor the thing to use Flask Blueprints & not be beautiful mind layout
Deploy the Local Carbon Footprinter container using the following command:
docker pull theboatymcboatface/local-carbon-footprinter:latest
Variable | Description | Default | Options |
---|---|---|---|
LOG_LEVEL | How many logs do you want? | INFO | DEBUG,INFO,WARNING,ERROR,CRITICAL |
APP_PORT | Which port should the web app use? | 3000 | Any Port Number |
TODO MachineType | What type of machine is this? | server | server, laptop, desktop |
TODO MachineName | Name of unique machine | Physical Host Name | Whatever you want |
If a route is not defined or a volume is not specified, a default volume will be created. The SQLite database 'db/carbon_footprint.sqlite' storing the footprint data does not require SSDs or any specialized storage infrastructure.
TODO Once local dev complete
Interested in contributing? Fantastic! I'm not quite there yet...
TODO
- Github Action Automations
- Good First Tasks
- Github Project
- Protect Branches
- Dependabot
- Supply chain things...
This project is open source. Contributions, suggestions, and improvements are welcome.
Open Source FTW. GPL-3