Improve application resilience with chaos engineering
summary
Claims that downtime costs enterprises up to $300,000 per hour (91% of surveyed firms) and roughly $10,000 per minute, then demonstrates resilience testing with AWS Fault Injection Service by running concrete experiments: terminate 50% of instances in an Auto Scaling group (Auto Scaling immediately launches a replacement), inject 200 ms latency to DynamoDB for 10 minutes (an alarm triggers a stop condition and the experiment rolls back), uses scenario libraries to simulate an availability‑zone outage, and shows a global safety lever to pause experiments during real incidents.
AI-written summary · about
chapters (11)
- 0:00Introduction: The Importance of Downtime
- 0:27Understanding the Cost of Downtime
- 1:47Challenges in Avoiding Downtime
- 3:31Building Resilient Systems
- 4:59AWS Resilience Lifecycle Framework
- 6:01Designing for Resilience
- 6:21Fault Isolation Boundaries
- 12:07Chaos Engineering and Resilience Testing
- 15:00Practical Example: Testing with AWS Fault Injection Service
- 22:11Advanced Scenarios and Safety Measures
- 25:22Conclusion and Final Thoughts
related talks
35mImprove resilience with automated chaos engineering
26mA proactive approach to resilience in modern apps
29mImprove Resilience with automated chaos engineering
31mPlanning for Failure using Chaos Engineering
42mChaos Engineering: Why Breaking Things Should be Practiced
57mChaos Engineering Primer
39mResilience & Chaos Engineering in the Cloud
26mBreaking Systems for Resilience