Journey to the Promised Land
summary
A concrete argument for planning operational work alongside features: a single typo once cost AWS $150M and downtime averages $5,600 per minute, so aiming for 99.9% = 42 minutes/month hides triage and deploy time. Teams here build pipelines and run kubernetes + terraform + helm but discover helm charts drop crash logs. Experiments like a shark tank meeting replace timers and forms, giving ten-minute cross-team pitches that noticeably cut firefighting and clarify pipeline requirements.
AI-written summary · about
chapters (36)
- 0:00Intro
- 0:30Downtime Hurts
- 1:44How long does it take to discover an outage
- 2:00The Promised Land
- 2:11Its a place on a map
- 3:01Different lenses
- 3:19tapestry of experiences
- 3:42Dont be down
- 5:44Take Charge of Your Journey
- 6:24How Many of You Can Draw an Owl
- 7:14Making Mistakes Automatically
- 7:58Practical Framework
- 8:26Why
- 8:37Culture
- 9:15Get your cape
- 9:53Where are you going
- 10:31Define your destination
- 10:56Culture of curiosity
- 11:16Learning as a currency
- 11:50What do we need
- 12:27Last year
- 13:55Making procedural changes
- 14:23bottlenecks warning
- 14:50priorities
- 15:05team rapport
- 15:51shoulder taps
- 17:03building stuff
- 17:29trying to solve
- 17:46taking a holistic view
- 17:58critical internal services
- 18:18sprawl
- 19:18Recap
- 19:36Where do we actually start
- 20:22Failing smart is faster
- 20:52Meet Jacquelyne
- 21:18Thank you
related talks
26mUptime, Mean-Time, and Ahead of Your Time
26mImprove application resilience with chaos engineering
29mMaking the Business Case for DevOps
29mBreaking Down the TofuLith. How We Reduced Lead Times for Multi-Tenant Infra Changes...
38mScaling Kubernetes Deployments Across +15'000 Namespaces
18mScaling DevOps in Complex Environments: Challenges, Failures, and Successes
40mAvoiding Pain When Operating in the Cloud
47mHow to win in software development with tighter feedback loops