Keynote: Observability in the Next Decade, Beyond Reactive Response
summary
A pager rings 91 times over two hours after a customer deployment illustrating repeated infrastructure breakage, not app tests, as the root cause. The talk claims a system converts incidents into availability signatures and builds a knowledge graph linking versions, configs, components and add ons to scan clusters and prevent repeats. It positions Datadog Prometheus Grafana as necessary but insufficient and describes a deployed product called check that implements a proactive prevention outer loop like a CVE for availability.
AI-written summary · about
related talks
12mKeynote: From Outage To Observability: Lessons From a Kubernetes...
33mKeynote: Observability is due for a version change; Are you ready for it?
13mLightning Talk: Observability-First DevSecOps: Building Resilient Multi-Cloud P...
09mKeynote: Scaling Observability Pipelines in the Enterprise
29mObservability; a gentle introduction
15mKeynote: The Observability Platform Engineering Advantage: From Zero-Code to Monitoring as Code
20mKeynote: Prometheus, Jaeger, and Envoy: The Observability for Today's Enterprises
06mKeynote: Beyond Observability: Intelligent Progressive...