How to Choose the Right Messaging Service for Your Workload
summary
Compares AWS bridges and queues by concrete tradeoffs: throughput quotas, Lambda concurrency behavior, cost per million messages and real-world scale. Recommends Kinesis on-demand for very high throughput (example customer ingests 500 million events a day at peaks of ~600k/sec) and using SQS as durable fallback. Notes retry and debugging differences—prefer Lambda destinations over DLQs for richer failure context—and warns about stream limits like 24-hour default retention that break replay unless extended.
AI-written summary · about
chapters (8)
- 0:00Intro
- 2:18Event-driven architecture
- 2:50Amazon EventBridge
- 6:16Scaling
- 14:24Costs
- 19:40Error handling
- 26:00Observability
- 38:47Outro
related talks
51mServerless architectural patterns
53mBottlenecks in Serverless Applications
30mApplication integration patterns for microservices
24mMigration from On-Prem Messaging System to The Cloud
40mSh... Fail Happens: Fail-aware Events Processing on AWS
22mS3 & Lambda - flexible pattern at the core of serverless applications
29mBuild highly scalable serverless applications on AWS
48mServerless design patterns