Keynote: Prometheus, Jaeger, and Envoy: The Observability for Today's Enterprises
summary
Claims that enterprises should adopt Prometheus, Jaeger, Envoy (Envoy baked into service mesh), OpenTracing and a central platform like GitLab to tame toolchain sprawl; cites $3.9 billion spent on DevOps last year and 87% of organizations disappointed as evidence. Recommends a hub-like model across the org, metrics-first alerts tied to SLOs, traces to locate failures, logs for forensic and compliance use, and instrumenting depth not breadth for critical transactions; service mesh helps as a hub but requires its own observability and deeper instrumentation.
AI-written summary · about
chapters (27)
- 0:00Introduction
- 0:16Presentation Overview
- 0:52Thesis
- 1:26The Conversation
- 2:29What is Observability
- 3:14Why does observability matter
- 3:55Speeding up cycle time
- 4:53The problem
- 5:03Integration complexity
- 5:42Tool chain crisis
- 6:48Microservices explosion
- 6:59Observability pillars
- 7:12Why these pillars
- 8:17The holy grail
- 8:59Tools
- 9:28Prometheus
- 10:34Flexible
- 11:03Traditional vs Ideal
- 11:48Building Observability Stackin Culture
- 13:29Defining Data
- 14:21Instrument for Depth
- 14:57Opinions Best Practices
- 16:02Alert Fatigue
- 16:55Service Mesh
- 18:00Recap
- 18:58Resources
- 19:16Thank you
related talks
20mKeynote: Service Meshes and Observability
45mOpen Source for Better Observability
28mOpen Source for Better Observability
29mObservability; a gentle introduction
31mCharting New Frontiers: The Evolution of Observability in Cloud Environments
27mOpening the Door to Observability
33mEffective observability in microservice architectures
33mWhat Developers Should Know About Observability