IoT Sensor Monitoring Demo (Azure SQL Edge + Docker + Grafana)🤖

This demonstrates real-time IoT sensor data visualization using:

• Azure SQL Edge (Docker container) as the database
• Python Flask API (sensor-api) to generate and serve fake sensor data
• Grafana OSS (Docker container) to visualize data
• Docker Compose for orchestration
• Environment variables (.env) for secure configuration

---

Features 🚢

• Generates fake sensor data for 3 devices every 5 seconds
• Stores data in Azure SQL Edge
• Provides /data API endpoint to expose the latest 100 records in JSON
• Grafana dashboard with:
  • Device Temperature (Time Series)
  • Device Humidity (Time Series)
• Safe handling of credentials via .env (no hard-coded secrets)

---

Project Structure 🦴

azure-grafana/
├─ docker-compose.yml
├─ .env # not committed
├─ sensor-api/
│ ├─ Dockerfile
│ ├─ requirements.txt
│ └─ app.py
├─ grafana/
│ └─ dashboards/
│   └─ sensor-dashboard.json

Flow

[Flask sensor-api]   →   INSERT every 5 seconds   →   [Azure SQL Edge DB]
        ↑                                                     ↓
        └────────────  Grafana requests /data  ───────────────┘

                         ┌───────────────────────┐
                         │      Grafana UI       │
                         │  (localhost:3000)     │
                         └───────────▲───────────┘
                                     │  HTTP (JSON)
                                     │  GET /data
                                     │
                       Docker Network│(bridge)
─────────────────────────────────────┼─────────────────────────────────────
                                     │
              ┌──────────────────────┴──────────────────────┐
              │                                             │
   ┌───────────────────────┐                     ┌────────────────────────┐
   │      sensor-api       │                     │    Azure SQL Edge      │
   │     (Flask App)       │                     │        Database        │
   │  - Exposes /data API  │                     │ - Stores sensor data   │
   │  - Generates fake     │                     │ - Persistent volume     │
   │    sensor data        │                     │ - Port: 1433           │
   │  - Inserts every 5s   │                     └───────────▲────────────┘
   │  - Port: 5000         │                                 │
   └───────────▲───────────┘                                 │
               │    SQL Query (SELECT / INSERT)              │
               │                                             │
               └─────────────────────────────────────────────┘

────────────────────────────────────────────────────────────────────────────
                            Docker Host Machine
────────────────────────────────────────────────────────────────────────────

---

How it works 🏃‍♀️

1. sensor-api/app.py:

• Uses environment variables to connect to Azure SQL Edge
• Creates SensorData table if it doesn't exist
• Starts a background thread that inserts fake data every 5 seconds
• Exposes /data endpoint returning latest 100 rows in JSON

2. Azure SQL Edge:

• Stores sensor data persistently in Docker volume
• Accessible from Grafana or local SQL clients

3. Grafana:

• Connects to sensor-api via JSON API plugin
• Dashboard visualizes temperature & humidity per device
• Refresh interval can be set (default 5s matches API data insertion)

Note 📕

• Data insertion only occurs while Docker containers are running
• Stopping docker-compose down halts insertion, but existing data persists in Docker volume
• For security, credentials are kept in .env and never in code

Further potential 🌕

• Connect real IoT sensors instead of fake data
• Add Grafana alerts for threshold values
• Integrate CI/CD pipelines for automatic deployment
• Use Docker Secrets or Azure Functions for more secure production deployments