Running Workloads in Kubernetes
summary
Shows how Kubernetes primitives map to concrete behaviors: use Deployment for stateless services with replicas=3 and rolling updates via kubectl so killing a pod immediately triggers replacement and traffic stays on healthy pods through a Service; use StatefulSet for stateful apps (demo: Zookeeper with 3 replicas, stable pod identities and volumeClaimTemplates so data survives pod recreation); use DaemonSet for one pod per node and Job for batch work (parallelism=3, completions=10). Helm and kubectl commands are used end to end.
AI-written summary · about
chapters (12)
- 0:00Introduction
- 1:16Why use Kubernetes
- 2:31Application patterns
- 3:38Kubernetes process
- 4:20Overview
- 5:18Deployment
- 6:39Demo
- 14:20Stapleset Demo
- 19:01Demons Demo
- 22:41Job
- 24:03Job Demo
- 27:04Recap
related talks
37mBuilding Stateful Workloads in Kubernetes
01h 02mKubernetes Patterns
02h 01mBuilding Reliable and Resilient Workloads in Kubernetes
41mUnderstanding Kubernetes in a Visual Way
39mIntroduction to Kubernetes
31m"Introduction to Kubernetes"
37mDeploying applications to Kubernetes: never send a human to do a machine's job
45mKubernetes Operability Tooling