TelSys: Real-Time IoT Telemetry Processing System

TelSys is a distributed backend system for ingesting, processing, and monitoring real-time telemetry data from IoT devices — designed to handle multi-metric classification, severity-based routing, and live device state streaming at scale.

Architecture

Each service is independently deployable and communicates asynchronously over Kafka.

Modules

• ingestion-api (8081) — Receives telemetry packets, applies per-device rate limiting via Redis, publishes to Kafka
• processing-worker (8082) — Consumes raw packets, classifies severity, persists to DB, updates Redis device state
• monitoring-api (8083) — REST + SSE API for querying telemetry, devices, and alerts
• alert-worker (8084) — Consumes critical/near-critical topics, creates deduplicated alerts
• simulator — Simulates 3 IoT devices sending telemetry every 3 seconds
• common — Shared models, DTOs, enums, and exceptions

Tech Stack

• Java 24, Spring Boot 3
• Apache Kafka (KRaft mode)
• PostgreSQL 16
• Redis 8
• Docker & Docker Compose
• Micrometer + Prometheus
• JUnit 5, Mockito

How to Run

git clone https://github.com/Flicko75/telemetry-system.git
cd telemetry-system
docker compose up --build

API Endpoints

Ingestion API (8081)

• POST /api/v1/telemetry

Monitoring API (8083)

• GET /api/v1/telemetry
• GET /api/v1/telemetry/{deviceId}
• GET /api/v1/telemetry/{deviceId}/range
• GET /api/v1/devices/{deviceId}/live (SSE)
• PATCH /api/v1/devices/{deviceId}
• GET /api/v1/alerts/{deviceId}
• PATCH /api/v1/alerts/{alertId}

Full API documentation available at:

• http://localhost:8081/swagger-ui.html
• http://localhost:8083/swagger-ui.html

Testing

• 70+ unit tests covering classification logic, service layer, and controllers
• Run with: mvn test