Cloud Chaos & Microservices Mayhem
summary
Cloud migrations still bite: a 2017 survey finds ~25% “zombie” servers and real bills can spike to $72,000 from a runaway Firebase job; JVM heuristics change below 2 CPUs/2GB and can cause long GC pauses when requests hop across ~100 microservices. Releasing cadence and FinOps gaps matter. Demonstrates OpenTelemetry for observability, uses Quarkus and a Node frontend with a Java backend, and proves Pact contract testing by generating a JSON contract and running a working verification demo.
AI-written summary · about
chapters (17)
- 0:00Intro
- 1:55Things you need to do well in 2022
- 6:50Tracing & observability
- 11:14Performance requirements
- 13:22Quarkus
- 14:18Performance optimization
- 19:36Takeaway
- 20:38Management
- 23:40FinOps
- 26:26Releasing
- 26:53Microservices
- 28:18Modularity
- 30:30Distributed monolith
- 33:20Test pyramid
- 37:53Demo
- 45:40Demo recap
- 47:12Outro
related talks
38mCloud chaos and microservices mayhem
48mJava in the Clouds: Pointless and Expensive or Productive and Must-Have?
31mEscaping operations hell
30mMigrating a monolith to Cloud-Native and the stumbling blocks
09mThe cloud native elephant in the room
14mKeynote: The Cost of Agility at Scale: It's All Fun and Games Until the Bill Comes
55mChaos Engineering for Cloud Foundry Platform
29mCost vs Stability in a Cloud Environment