https://github.com/aaronpowell/sunshine

Sunshine is an application I use to monitor my solar panels, and is a demo of building an IoT project using Azure.
https://github.com/aaronpowell/sunshine

azure-functions azure-iot-edge azure-iothub azure-storage-table fsharp iot

Last synced: about 1 month ago
JSON representation

Sunshine is an application I use to monitor my solar panels, and is a demo of building an IoT project using Azure.

• Host: GitHub
• URL: https://github.com/aaronpowell/sunshine
• Owner: aaronpowell
• License: mit
• Created: 2019-03-25T22:02:05.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2020-12-10T04:56:13.000Z (about 4 years ago)
• Last Synced: 2024-11-17T19:47:12.321Z (about 2 months ago)
• Topics: azure-functions, azure-iot-edge, azure-iothub, azure-storage-table, fsharp, iot
• Language: F#
• Homepage:
• Size: 2.62 MB
• Stars: 9
• Watchers: 3
• Forks: 3
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        


    

  
Sunshine - My Home Grown IoT Project

  
Sunshine is a project I started to learn how someone would go about building an IoT project. My use case was to connect to my solar inverter and export the data to store where I could produce my own reporting against it.



## How it works

The project is made up of 3 components:

* Downloader

  * .NET Core application written in F#

  * Deployed as a Docker image

  * Pulls the data from the inverter and pushes it to [Azure IoT Hub](https://azure.microsoft.com/services/iot-hub/?WT.mc_id=javascript-0000-aapowell)

* Functions

  * [Azure Functions](https://azure.microsoft.com/services/functions/?WT.mc_id=javascript-0000-aapowell) for processing the data, written in F# against the v2 runtime

  * Listens to a set of [Event Hubs](https://azure.microsoft.com/services/event-hubs/?WT.mc_id=javascript-0000-aapowell) for incoming messages

  * Writes to [Table Storage](https://azure.microsoft.com/services/storage/tables/?WT.mc_id=javascript-0000-aapowell)

* Mock API

  * A fake implementation of my inverter API so I can dev when I'm not at home

  * Implemented using Saturn, a F# web framework

The Downloader runs in a Docker container and will either talk to the Mock API (local dev) or the inverter (production) and push the data up to [Azure IoT Hub](https://azure.microsoft.com/services/iot-hub/?WT.mc_id=javascript-0000-aapowell). IoT Hub will then route the data to one of several [Event Hubs](https://azure.microsoft.com/services/event-hubs/?WT.mc_id=javascript-0000-aapowell) which the Functions are subscribed to for processing and storage in [Table Storage](https://azure.microsoft.com/services/storage/tables/?WT.mc_id=javascript-0000-aapowell).

Deployments are orchestrated using [Azure Pipelines](https://azure.microsoft.com/services/devops/pipelines/?WT.mc_id=javascript-0000-aapowell) to build the images, push to [Azure Container Registry](https://azure.microsoft.com/services/container-registry/?WT.mc_id=javascript-0000-aapowell) and then use [Azure IoT Edge](https://azure.microsoft.com/services/iot-edge/?WT.mc_id=javascript-0000-aapowell) to deploy onto either my test device or my production device.

You'll find more information on the process on my blog through the series [Home Grown IoT](https://www.aaron-powell.com/posts/2019-05-30-home-grown-iot-prologue).

## Acknowledgements

Logo: solar by iconsmind.com from the Noun Project