Designing and Deploying Scalable and Secure Applications on Kubernetes
summary
Practical checklist: use Deployment vs StatefulSet correctly (misused for SQL Server in examples), separate stateful from stateless, require at least two replicas, set resource requests/limits so the Horizontal Pod Autoscaler and Cluster Autoscaler behave, add readiness/liveness/startup probes and honor SIGTERM with ~30s prestop work, use Secrets and sealed Secrets by bitnami for Git workflows, emit Prometheus metrics and OpenTelemetry traces, apply PodDisruptionBudgets, network policies and pod affinity; the presenter counts 16 principles and reports an old predictive autoscaler built ~10–11 years ago that nobody adopted.
AI-written summary · about
chapters (6)
- 0:00Start
- 10:11Separate secret from non-secret configuration
- 14:18Enable automatic scaling to ensure capacity management
- 19:49Use probes for automatic failure detection and recovery
- 30:34Prepare your application for observability
- 33:26Set Pod resource requests and limits appropriately
related talks
19mKube Me This! Kubernetes Ideas and Best Practices
35mAutoscaling Kubernetes Deployments: A (Mostly) Practical Guide
39mKubernetes is awesome – but …
59mA DevOps State of Mind: Managing Microservices + Databases with Kubernetes
28mRunning Microservices on Kubernetes – Best Practices
45mKubernetes Resiliency
42mMastering Kubernetes: Challenges, Strategies, and Best Practices
35mTo Infinite Scale and Beyond: Operating Kubernetes Past the Steady State