LagWatch

  • Built a lightweight host observability service in Go using a concurrent collector/aggregator pipeline with goroutines and buffered channels.
  • Captured CPU, memory, disk, and network telemetry with low runtime overhead for real-time lag diagnosis.
  • Exposed JSON snapshot APIs for quick out-of-band checks and a Prometheus-compatible endpoint for historical analysis.
  • Integrated Prometheus and Grafana dashboards to convert raw metrics into actionable bottleneck and capacity signals.