Better Reliability Through Observability and Experimentation
summary
Shows how observability plus controlled experiments shorten time to detect (TTD) and harden systems using OpenTelemetry/OpenTracing, Gremlin-style fault injection, and game days. Cites a quick survey with 97 responses, teaches the four golden signals—latency, errors, traffic, saturation—and prescribes science: baseline metrics, hypothesis, design experiment, start at 5% blast radius then scale, set abort conditions, document results, and use experiments (dev → staging → prod) to find failure modes before customers do.
AI-written summary · about
chapters (4)
related talks
53mObservability Driven Development
38mOff the Beaten Path: Moving Observability Focus from Your Service
31mNo-Nonsense Observability Improvement
49mImproving Observability of Distributed Systems
58mCultivating Production Excellence
26mYou Can’t Fix What You Can’t See: Tales from the Observability Frontier
13mBeyond observability: Using reliability scores to drive results
39mShifting Left on Production Excellence with Observability