Improve Resilience with automated chaos engineering
summary
Claims that automating chaos lets teams run many more experiments and continuously verify assumptions over time rather than one‑offs. Shows concrete builds: a serverless scheduler (CloudWatch/EventBridge + Lambda) that runs AWS Fault Injection Simulator templates, an event‑triggered flow for autoscaling events, and a CI/CD stage using Step Functions to gate deploys. Demonstrates experiments running and an alarm‑triggered stop that fails the pipeline, proving the automation works with stop conditions in place.
AI-written summary · about
chapters (16)
- 0:00Automated chaos intro
- 1:17What is chaos engineering
- 3:45Manual chaos use cases
- 5:52Manual chaos limits
- 9:26Need for automated chaos
- 10:52Scientific automated approach
- 11:14Scheduling chaos tests
- 13:00Demo scheduled chaos
- 16:15Event-triggered chaos
- 16:59Demo event chaos
- 20:33Chaos in CI/CD pipelines
- 21:16Demo pipeline chaos
- 23:52Handling failed pipeline chaos
- 26:45Three automation approaches
- 27:50Key takeaways
- 28:11Q&A
related talks
35mImprove resilience with automated chaos engineering
26mBreaking Systems for Resilience
27mChaos Engineering: The Path to Reliability
34mChaos engineering for serverless
42mChaos Engineering: Why Breaking Things Should be Practiced
26mFrom experimentation to continuous verification
18mChaos Engineering: Break It On Purpose
26mImprove application resilience with chaos engineering