What happens when DynamoDB explodes? A practical guide
summary
Explains whether a DynamoDB outage deserves an immediate wake‑up by showing concrete diagnostics and remediation steps: how to map business impact to failing requests, DynamoDB throttling at 2:32 AM, tradeoffs between CloudWatch, OpenTelemetry/Jaeger and home‑built tracing, and what event correlation looks like in practice. Shows a live Lumigo demo that builds a virtual stack trace, exposes the exact failing payloads and retries, and demonstrates how many invocations and retries change the diagnosis.
AI-written summary · about
chapters (7)
- 0:00DynamoDB failures overview
- 1:28Managed services troubleshooting
- 3:51Distributed tracing need
- 7:42Open source tracing tools
- 10:30Cloud-native serverless monitoring
- 13:18Lumigo observability practice
- 20:17Key takeaways
related talks
33mBuild for scale with Amazon DynamoDB developer features
43mBest practices for continuous testing and observability
39mBuilding Resilient Serverless Applications
36mObserving all the Serverless Things
26mHow to survive a Lambda-induced data tsunami in AWS
47mEvent-driven Serverless Architecture with DynamoDB
37mDynamoDB in an EDA World
26mImprove application resilience with chaos engineering