Cloud Computing Made Better: An AWS View on Serverless Computing
summary
Serverless adoption has exploded: over half of new applications built inside Amazon, driving trillions of invocations per month and some customers at tens of thousands of concurrent invocations. AWS maps the Berkeley predictions to reality: limits on ephemeral and persistent storage, communication/coordination unpredictability, function runtime and startup constraints, and multi-tenant security remain. Remedies include on‑demand DynamoDB, managed containers and microVMs, snapshotting and specialized hardware, higher‑level workflow planners, and research into ML to infer workload signals rather than require verbose user specifications.
AI-written summary · about
chapters (4)
related talks
31mThe Evolution of Serverless: Past, Present and Future
32mThe Serverless Tidal Wave
42mNo Server Is Better Than No Server: Building Blocks for Serverless Architectures
47mAdvanced Serverless Architectural Patterns on AWS
31mAccelerate business innovation with serverless applications
01h 00mServerless - reality or BS - notes from the trenches
50mServerless Architecture Patterns and Best Practices
28mThe Future of Serverless