The Cloud Native Journey: Navigating the Stages of Maturity in Observability
summary
Presents six stages of observability maturity with specific practices: start with availability dashboards and restart playbooks, add PromQL alerts (example: memory free <5%), centralize logs with Fluentd + OpenSearch, instrument apps with OpenTelemetry and Prometheus, and use long-term stores like Thanos/Cortex and Jaeger for traces. Shows a real case of hosts at <50% memory for two months (scale to save money) and extrapolates savings across 40K hosts. Recommends right-size memory and use OpenCost for capacity/cost optimization.
AI-written summary · about
chapters (8)
- 0:00Introduction to observability
- 3:56Stage one dashboards and alerts
- 7:54Stage two alerts and monitoring
- 12:55Stage three proactive monitoring
- 17:40Stage four application instrumentation
- 22:42Stage five optimization and cost
- 25:10Tools and resources overview
- 29:48Q&A and closing
related talks
16mBuilding Observability into Cloud-Native Applications
32mThe Role of Observability in Cloud Native Environments
31mCloud Native Observability
16mShrinking the Observability Bill
31mCharting New Frontiers: The Evolution of Observability in Cloud Environments
44mGrafana Labs Customers: What We've Learned Building Observability at Massive Scale
21mtalks about mastering Observability for DevOps
29mObservability; a gentle introduction