Architecting Production Development Environments for Reliability and Disaster Recovery

Learn how Meta architects its development environment for reliability in this technical conference talk that explores the infrastructure powering their developers' daily work. Dive into Meta's software architecture, examining both virtual machine devservers and On Demand containers while understanding the mechanisms that keep these systems updated, reliable and available. Explore key concepts including disaster recovery, production engineering practices, and strategies for maintaining system stability during maintenance workflows and outages. Gain practical insights into user-facing development workflows, disaster preparation, runbook creation, communication protocols, and live migration techniques. Discover valuable lessons learned from past incidents and how they've shaped Meta's approach to building robust development environments at scale.