Production Readiness Strategies for an Automated World
summary
Argues that production readiness is about checklists, automation, and organizational practices more than raw tech; demonstrates concrete items to demand before deploy: reproducible builds, metrics and TP99 health, tracing tokens, automated failover and geo tests, and documented runbooks. Shows how simple echo service yields a complexity factor ~10 while a database multiplies that to ~30+, and highlights practical tooling: Go services, Nomad job meta fields for owners/docs, Consul DNS health failover, and the rule that backups are useless restores matter (zero-size is an alarm).
AI-written summary · about
chapters (10)
- 0:00Introduction and speaker background
- 8:47Why checklists for production readiness
- 12:02Organizational prerequisites for production
- 19:54Service checklist overview
- 34:07Defining outages and severity levels
- 41:12Managing single points of failure
- 46:25Architectural limits and logging
- 51:02Backup strategies and acceptable loss
- 55:35Organizational continuity over tech
- 58:03Q&A and closing remarks
related talks
20mAre you ready for production?
43mGetting Your System into Production and keeping It There
46mPracticalities of Productionizing Distributed Systems
42mProduction readiness in Azure: A practical guide
58mCultivating Production Excellence
26mMission Critical Mindset Building a Culture of Resilience in DevOps Teams
42mAutomate all the things
41mLessons from Mission-Critical Systems in Fast-Moving Industries