AI and Chaos Engr: Smarter Failure Testing for Resilient Systems
summary
Claims that AI auto-generates and prioritizes chaos scenarios, ties into observability (EC2, Prometers, OpenTelemetry) and injects faults via tools like chaos monkey Gremlin Litmus and Chaos Toolkit, using ML for anomaly detection and reinforcement learning for scenario planning. Case studies state Netflix uses predictive analysis and AI simulations to sustain ~99.9% availability and a Gremia deployment cuts query time by 60% and avoids major outages. Implementation notes include starting small, keeping a human in loop, and measuring availability latency MTTR.
AI-written summary · about
chapters (10)
- 0:00Introduction and Agenda Overview
- 1:03Understanding Chaos Engineering
- 2:09Evolution and Limitations of Chaos Engineering
- 2:52AI's Role in Enhancing Chaos Engineering
- 3:48AI-Driven Chaos Engineering Workflow
- 4:45Tools and AI Integration in Chaos Engineering
- 6:00Case Studies and Real-World Applications
- 6:56Benefits and Challenges of AI in Chaos Engineering
- 7:44Best Practices and Future Trends
- 8:26Conclusion and Final Thoughts
related talks
39mSmart Chaos : Leveraging Generative AI
43mTaming the Unpredictable: Reliability in Chaos
28mAI for Everything After Code
32mQuality engineering in the AI age: From testing AI to leveraging GenAI
31mResilience in the Uncharted AI Landscape
26mAI that is gold does not glitter - using intelligence in Modern Infra
05mAutomating Cloud-Native Infra with Intelligent Tools
44mHow AI is Exposing the Existential Crisis of Software Testing