Ignite - What I Learned From a Dress an On-Call Nightmare
summary
BuzzFeed experiences a viral spike tied to the Dress that drives 670,000 active users and breaks the site: a NoSQL-backed voting app autoscaling handles votes fine while the legacy Perl CMS and front-end stack fail under load. Ops are racking hardware in a physical datacenter, fighting hot MongoDB clusters, memcache limits, and an architecture that deliberately avoids some AWS services. The team implements rapid mitigations, restores voting throughput, and documents the human toil and technical debt exposed.
AI-written summary · about
related talks
06mWhat I Learned From A Dress, an On-Call Nightmare
20mWhat I Learned From A Dress, an On-Call Nightmare
06mIgnite: Migrating A Legacy App to Modern Infrastructure
28mReliability Lessons from Running High Traffic Services for More Than 10 Years
04mMark Imbriaco interviewed at Velocity 2011
36mX-Factor Voting Platform Under the Hood
25m"The website's down!" Stories and lessons on keeping your website up.
06mIgnite: Moving from regular sysadmin to DevOps