Harnessing Chaos Engineering to Build Robust Systems
summary
Claims that OpenTelemetry reduces vendor lock‑in and standardizes tracing, and that Netflix Uber and Airbnb run the stack in production using Prometheus + Grafana for metrics, Ager exporters for traces and elastic stack for logs. Concrete outcomes: faster MTTR, autoscaling driven by metrics, and rising costs as telemetry volume multiplies. Demos of instrumentation cadence (every 2–3 weeks to quarterly deploys) and AI anomaly detection plus eBPF for deeper observability are presented as working directions.
AI-written summary · about
chapters (10)
- 0:00Introduction to Cloud Native Observability
- 0:49Overview of Presentation Topics
- 1:17Defining Cloud Native Observability
- 2:20Importance of Observability
- 5:31Key Observability Tools and Technologies
- 8:46Best Practices for Cloud Observability
- 9:55Real World Use Cases
- 10:33Current Challenges in Observability
- 11:11Future Trends in Observability
- 12:15Conclusion
related talks
22mSeeing Through the Chaos of Modern Systems
46mObservability Ecosystem in Kubernetes
42mDistributed Monitoring: How to understand the Chaos
38mThe State of OpenTelemetry
14mObserving the Future
50mOpen Telemetry: The Rise of Open Standards in Observability
38mChaos Engineering
20mObservability and Open Telemetry