Thinking Asynchronously
summary
Shows how a synchronous translation pipeline is rewritten to store first, process later using Amazon API Gateway service integrations, DynamoDB, DynamoDB Streams and EventBridge rules to trigger tiny Lambda jobs that call Amazon Translate and Polly, drop MP3s to S3, and write results back to DynamoDB. Claim: client wait drops from seconds to nearly instant because compute runs asynchronously; demo is built and works; Lambda Destinations plus a periodic error-correction Lambda provide automatic retries and observability.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 2:37Application elements
- 5:36Thinking asynchronously
- 11:40Making this work on AWS
- 14:21Other "storage first" options
- 15:01Event data
- 18:07Resulting Data
- 23:23Asynchronous demo
- 31:13More reliability - Lambda Destinations
- 32:46More reliability - Error correction
- 34:25Extending via the pattern
- 37:05Wrap up
related talks
22mS3 & Lambda - flexible pattern at the core of serverless applications
31mA day in the life of a serverless developer
30mEvolving Serverless Architectures
47mEvent-driven Serverless Architecture with DynamoDB
41mUnlocking Faster & Efficient Data Processing w/ Serverless
51mServerless architectural patterns
33mAWS Lambda - Logic in the Cloud without Back-end
31mAccelerate business innovation with serverless applications