Lightning Talk: Mission: Impossible - Building Self-Healing Disaster Recovery With...
summary
A production system uses OpenFeature as a control plane to orchestrate dynamic failover across three regions, processing 1k flag evaluations/s and cutting RTO from 25 minutes to 125 seconds RTO while lowering infrastructure incident risk by 90% and surviving 15 regional incidents with no lost transactions. An infra gateway controller tracks database health, syncs global flag state (radius), enforces millisecond traffic shifts, and triggers automatic warm‑hot‑cold traffic transitions with zero manual intervention.
AI-written summary · about
related talks
47mBuilding Resilience
22mFailing Over without Falling Over
25mAutomating Disaster Recovery: The Ultimate Reliability Challenge
14mExpect the unexpected: Building resilience with AWS
16mFrom System TAYE7 to High Availability: What, Why, How?
16mResilient by Design: Architectures for Crisis Response & Recovery
14mThe Multi‑Workload Challenge: Predictable Failover
25mKeynote: Infrastructure - Introduction