Build your next microservices application with modern AWS services
summary
A small demo application is built with a Node.js front end, a Lambda parser that massages input, SQS for queuing, EKS worker pods that persist to DynamoDB, all provisioned by CloudFormation stacks. Live load breaks the fragile demo (pods restart 16 times), observability via New Relic exposes CPU/memory hotspots across three nodes and traces slow requests, and an optimizer surfaces $449/month reclaimable from right‑sizing. The concrete claim: modern AWS services reduce ops burden but require end‑to‑end telemetry to diagnose scale failures.
AI-written summary · about
chapters (38)
- 0:00Introduction
- 1:10About the speaker
- 3:52About Fleet Complete
- 4:53Agenda
- 5:43Why
- 7:10Transformation
- 8:41Walk stage
- 10:32transformational journey
- 12:49best options
- 14:11cost efficiencies
- 17:40game time
- 18:20architecture
- 19:13demo
- 22:33key takeaways
- 23:58New Relic foundation
- 26:05Why did it break
- 26:47Still up
- 27:18Manage the service
- 28:00Areas of concern
- 28:29Your workload
- 29:13Whats on fire
- 29:55Single pane view
- 30:38Entity list
- 32:05telescopic view
- 32:58log view
- 33:38error
- 34:05context events
- 34:39distributor traces
- 35:38use the wrong browser
- 36:11reclaim wasted compute
- 36:30in application optimized
- 36:50saving money
- 37:14compute optimized
- 37:58export report
- 39:07smart phone
- 40:03takeaways
- 41:45failure mode analysis
- 46:00Closing thoughts
related talks
53mLook Ma! No servers!
02h 37mServerless Deep Dive
50mBuild, Deploy, Debug and Tune Serverless Microservices
43mDesigning and building resilient backend APIs in the cloud
04mServerless applications, Microservices, AWS
51mTaking Serverless to the Next Level
49mMonitoring serverless applications on AWS: the monitoring monster becomes your friend.
28mServerless Backend for Frontend on AWS