Build highly scalable serverless applications on AWS
summary
Describes a serverless stack that uses Amazon API Gateway, AWS Lambda, Aurora/DynamoDB, RDS Proxy, ElastiCache/DAX, CloudFront, SQS/SNS/EventBridge/Kinesis and Power Tuning to hit tens to 100,000 requests per second and support millions of users. Gives the Lambda concurrency formula and warns default account limits (API Gateway 10k RPS, Lambda concurrency 1,000) and the hard 29 seconds timeout on API Gateway. Playvox reports ~3x throughput growth in 12 months, 3x cost efficiency, and lessons: watch quotas, add caching early, use async queues and multi-account isolation.
AI-written summary · about
chapters (12)
- 0:00Introduction and overview
- 3:00Basic serverless architecture
- 4:07Synchronous request scaling
- 8:35Optimizing Lambda functions
- 11:15Database and caching strategies
- 14:13Asynchronous messaging patterns
- 17:51Multi-account architecture
- 19:40Recap and key takeaways
- 20:02Playvox journey to serverless
- 23:18Lessons learned and challenges
- 25:30Future plans and improvements
- 27:20Closing and resources
related talks
51mArchitecting for Hyperscale with AWS: Navigating Real-world Serverless Challenges
31mAccelerate business innovation with serverless applications
42mServerless Tales from the Trenches
39mFrom 0 to 60,000 RPS in 1 Minute with AWS Serverless Services
36mTransforming SaaS with Serverless Architectures
21mBeyond the Basics: Production Serverless Patterns for Extreme Scale
31mFullstack Serverless at Scale
35mWhy serverless will revolutionize your software process