Designing for Failure - Scaling Uber by Breaking Everything

Explore the challenges and strategies of scaling a high-availability system in this conference talk from YOW! 2015. Learn how Uber's engineering team tackles the complexities of maintaining reliability while rapidly expanding their business and workforce. Discover the shift from a small service with few components to a global operation with hundreds of services, and how this transition impacts both technical and cultural aspects of the organization. Gain insights into Uber's architecture and its approach to handling various failure scenarios, including regular failure testing of new systems and databases. Examine real-world outages and their influence on system design. Understand the importance of designing for failure and making failures more common to improve overall system resilience. Ideal for software engineers, architects, and anyone interested in building scalable and reliable systems for rapidly growing businesses.