Understanding Progressive Collapse: How To Avoid A Cascading Failure

Why watch

Sam Newman applies civil engineering's progressive collapse concept to distributed systems, analyzing real failures including the Ronan Point tower collapse, AWS us-east-1 DNS outage, and the Used Car Site thread saturation incident. The talk provides three architectural mitigations: reducing hazards through load shedding, strengthening components via multi-region redundancy, and decoupling interconnections, aimed at senior architects and SRE engineers.