An IoT data collection and visualization platform built with Node.js and AWS. IoT devices push telemetry data via a REST API for real-time monitoring and historical analysis.
IoT devices send telemetry via HTTP POST to an Express.js API running on an EC2 instance. Data is stored in MongoDB with time-series collections for efficient range queries. A Vue.js dashboard displays real-time sensor readings via WebSocket and historical charts using Chart.js. The system supports multiple device types with a flexible JSON schema.
When multiple devices reported simultaneously, the single-threaded Node.js event loop struggled under high concurrency. The fix was to introduce a Redis-backed message queue to decouple ingestion from persistence, along with connection pooling for MongoDB.
Without data retention policies, the database grew rapidly. MongoDB's TTL indexes were configured to automatically expire data older than 90 days, and aggregation pipelines were used to downsample older records into hourly summaries.