It’s Easy to Say Serverless
summary
Describes using Function-as-a-Service (serverless) to run an event-driven photo pipeline: uploads trigger parallel image processing and the cloud autoscaling removes manual server provisioning. Gives production metrics (peaks ~1,500–2,500 calls/minute; a photo site with 400k photos, 130k photographers, 3,003,000 images), reports fewer servers and faster deploys, and discusses real issues and mitigations — cold starts, vendor lock-in, longer call chains, and patterns like API proxy, shared libraries, and cross-region disaster recovery.
AI-written summary · about
chapters (8)
- 0:00Serverless introduction
- 2:36Serverless functions
- 8:19Photo platform case study
- 16:50Dependency management
- 23:10Disaster recovery
- 25:25Practical experiences
- 30:24Economic benefits
- 32:28When to use serverless
related talks
58mServerless - the brief past, the bewildering present, and the beautiful (?) future
52mServerless landscape beyond functions
50mA hitchhicker's guide to Serverless computing
20mWhy Serverless Architecture is Changing Software Development
30mThe Hitchhiker's Guide to the Serverless Galaxy
42mNo Server Is Better Than No Server: Building Blocks for Serverless Architectures
20mThe Serverless First Mindset
51mServerless - a backend thing that gives superpowers to frontend developers