This project runs on a raspberry Pi with a temperature/humidity sensor and a DigitalOcean droplet running MongoDB and the MongoDB BI tool.
To start, pi_code contains the code to run temperature and humidity gathering on the Pi and reports that data to the MongoDB droplet, storing temperature and time data. We'll also run our cubejs instance off of the Pi. The code for this is contained in cube-instance
To start, create a MongoDB droplet (this tutorial helps) and use ssh port forwarding on your pi to map your localhost port 27017 and whichever port you plan to use for the BI tool (I used 27018) to your droplet.
Then, install the MongoDB BI tool on the droplet using the tutorial in the documentation. Configure and run it as a daemon.
Install Node.js and npm on your Pi, then run npm i in the pi_code directory. Use sudo raspi-config to enable I2C interface.
Run the pi sensor code with sudo node index.js
Go into the cubejs-instance folder and run npm i again, then start the cubeJS instance with npm run dev.
Go to the CubeJS instance at localhost:4000 and look through the Schema and set up some charts! You can pivot on Time and Temps.