EngineerExperiences

Part 3 of 12 ESP32C++ReactInfluxDBGrafana

Pulzaura System Architecture

Pulzaura is designed around an end-to-end telemetry pipeline streaming real-time biosignals from embedded wearable sensors to web dashboards and time-series analytics databases.

Pulzaura telemetry path from the wearable to the backend, dashboard, database and Grafana ESP32 wearable MAX3010x, AD8232 Backend (C++) oatpp + BlueZ BLE GATT notify Frontend (React 19) Vite, live graphs InfluxDB storage continuous persistence Grafana dashboards auto-provisioned, 5s refresh WebSocket (JSON frames) HTTP line protocol dashboard queries
One BLE link carries every reading; after that the data stays local, so nothing depends on a third-party service.

Architectural Tenets

  1. Low-Latency Streaming: Sensor packets bypass disk write cycles on the front path, routing directly from the BlueZ Bluetooth subsystem to client WebSockets.
  2. Resilient Long-Term Persistence: Every packet received is buffered and written to InfluxDB in batches using Influx Line Protocol.
  3. Decoupled Visualization: React delivers responsive, sub-50ms live waveform rendering, while Grafana delivers aggregate historical trends and multi-hour session analytics.