Applying Observability to the Internet of Living Things (IoLT)
summary
A virtual cat called Meowy is built in Go, instrumented with OpenTelemetry and exported via OTLP to an Elastic/ELK backend and Kibana, with a batch span processor and 5-second metric exports. Code records counters (escape attempts = 5 observed), histograms for nap-location clusters, traces for each state span, and a computed average MPH (meows per hour). Dashboards show heatmaps of naps, time series of poops and food consumption, and alerts can auto-order food.
AI-written summary · about
chapters (14)
- 0:00IoLT and observability basics
- 2:33Observability beats debugging
- 3:38OpenTelemetry standard
- 5:28Virtual cat demo
- 7:15Setting up OpenTelemetry in code
- 8:21Metrics and distributed tracing
- 10:10Virtual cat state machine
- 11:59Live virtual cat demo
- 12:40Observability dashboards
- 14:29Key Elastic visualizations
- 15:53Service maps and features
- 16:58Lessons from observable systems
- 19:10Journey to observability
- 19:52Conclusion
related talks
55mCloud-Native Observability with OpenTelemetry
48mAPI Observability by Design: Building Systems That Tell Their Own Stories
34mOpenTelemetry for Dummies: Instrumenting Go Apps
52mGraphQL Observability with Elastic and OpenTelemetry
01h 00mObservability made painless: Go, Otel & LGTM stack
37mObservability For You And Me With Opentelemetry
20mObservability and Open Telemetry
26mPlug-and-Play Observability: OTel Auto-Instrumentation for LLMs in OpenSearch