Journeys To Cloud Native Architecture: Sun, Sea & Emergencies
summary
Clients try to launch cloud‑native stacks—Kubernetes, Docker, CI/CD—but often take on 40–50 microservices at once and hit missing logs (about half), platform backpressure, and a 4x performance slowdown versus Docker Compose. Recommendations are concrete: start with 1–2 marker services, outsource commodity ops to managed Kubernetes (EKS/AKS/GKE), add observability (Zipkin, distributed tracing, Honeycomb), use declarative tools (Terraform, GitOps), run security checks (kube‑hunter, kube‑bench) and chaos tooling (Gremlin/chaostoolkit) and focus on automated recovery.
AI-written summary · about
chapters (11)
- 0:00Introduction
- 0:32Feeling pain Learning from failures
- 3:42Overview - The Dream
- 6:05"Going Cloud-Native" is an approach to build and run applications capable of exploiting and taking advantage of modern, scalable cloud like delivery techniques and tools in order to meet business objectives!
- 8:26Tackling Too Much Too Quickly (Sunburn)
- 15:43Relying on a Surface Level Understanding (Caught In Deep Water)
- 16:32Distributed Systems are hard. The devil is in the detail And the detail matters
- 22:19Other War Stories ... DRY Dependency Hell • The Singularity Fallacy
- 36:19Ignoring Security (Living Dangerously)
- 47:27Summary & Principles for Success
- 48:08Don't take on too much too quickly • Acquire enough expertise to journey safely • Choose your Tools wisely • Engage don't enrage Security . Anticipate, Plan and Actively Test for Failure
related talks
27mCloud Native Landscape for Node.js Developers
18mArchitecting Resilient Cloud Native Applications: A Practical Guide to Deployment and Runtime
08mThe Evolution of Microservices Architecture in Cloud-Native
28mCloud Chaos and Microservices Mayhem
28mRunning Microservices on Kubernetes – Best Practices
54mPanel Discussion
37mCloud Anatomy 101: Breaking Down Cloud-Native
35mKubernetes journey