Kubernetes Resiliency

Explore Kubernetes resiliency in this comprehensive NDC Oslo 2023 conference talk. Delve into infrastructure considerations, including single points of failure, availability zones, and node pools. Examine critical Kubernetes components such as etcd, API server, and kube-scheduler. Learn about application-level resilience strategies, including pod disruption budgets, resource requests and limits, and liveness and readiness probes. Gain insights into ensuring cluster and application resilience, covering topics like networking, service discovery, rate limiting, circuit breaking, and autoscaling. Discover essential Kubernetes web resources and production best practices to enhance your understanding of building robust, scalable systems on the Kubernetes platform.

Introduction
Agenda
Infrastructure
Networking
Availability Zones
Node Pools
Infrastructure Level
Kubernetes Components
Service Discovery
Rate Limiting and Circuit Breaking
Metric Server
Horizontal Pod Autoscaling
Event Driven Autoscaling
Pod disruption budgets
Resource requests and limits
Quality of services
Liveness Readiness probes
Kubernetes web resources
Kubernetes production best practices
CPU limits