What Is Kubernetes? A Visual Guide to Its Core Concepts and Real-World Uses

Kubernetes Overview

Kubernetes (k8s) is an open‑source container orchestration platform that manages containerized applications across multiple hosts. It provides rapid deployment, scaling, upgrading, and maintenance, along with service discovery, Layer‑4 load balancing via Service and Layer‑7 load balancing via Ingress. Originating from Google in 2014, it builds on extensive production experience and a large community.

Key Advantages

Excellent user experience : Web UI enables quick deployment, upgrade, scaling, and rollback of applications.

Intelligent pod scheduling : Pods are placed on nodes with sufficient resources, balancing workloads and spreading replicas.

Flexible deployment environments : Supports public, private, hybrid clouds, VMware, virtual machines, and bare metal.

Robust authentication, authorization, and auditing : Fine‑grained access control and real‑time operation logs simplify troubleshooting.

Strong scalability : Automatic cluster expansion or contraction matches business demand.

Disaster‑recovery and alerting : Built‑in backup and failover mechanisms enable near‑instant switchover.

Windows node support : Allows deployment of Windows containers alongside Linux workloads.

Practical Application Scenarios

1. Log Management (EFK Stack)

Kubernetes can deploy the Elasticsearch‑Fluentd‑Kibana (EFK) stack using a DaemonSet. Fluentd runs on every node, automatically collecting logs from new nodes when the cluster scales, and the logging pipeline can be horizontally scaled to handle growing data volumes.

2. DevOps Pipelines

By integrating with CI/CD tools, Kubernetes provides multi‑cluster management, immutable Docker images for environment consistency, and automated build, test, and release cycles. It enables continuous integration, delivery, and deployment across development, testing, and production clusters.

3. Microservices Architecture

Features such as Service Mesh, ConfigMap, Horizontal Pod Autoscaler (HPA), and rolling updates support service discovery, unified configuration, elastic scaling, and zero‑downtime deployments for microservice workloads.

Core Characteristics

Portability : Runs on public, private, and hybrid clouds (multi‑cloud).

Extensibility : Modular, plug‑in‑friendly, and composable architecture.

Self‑healing : Automatic pod placement, restarts, replication, and scaling.

Detailed Feature Set

Multi‑tenant network isolation : Supports CNI plugins (e.g., Calico, Flannel) and network policies to isolate namespaces.

High availability : Solutions such as keepalived + NGINX or keepalived + HAProxy provide failover for API servers and other control‑plane components.

Persistent storage options : Local storage (emptyDir, hostPath), network storage (iSCSI, NFS, CIFS), distributed storage (GlusterFS, Ceph, CephFS), and cloud disks (e.g., Azure Disk).

Update and rollback strategies : Supports blue‑green, canary, and rolling updates via controllers and services.

Elastic scaling : Horizontal Pod Autoscaler adjusts replica counts based on traffic, achieving near‑instant scaling.

Comprehensive runtime capabilities : Process coordination, storage mounting, secret distribution, health checks, replica management, load balancing, monitoring, logging, debugging, authentication, and more, delivering PaaS‑level conveniences on top of IaaS.

Kubernetes Architecture

The system follows a master‑node model. Masters (typically 3 or 5 for HA) run the control‑plane components, while worker nodes host the workloads.

Core Components

Master Node

kube-apiserver : Central API entry point handling authentication, authorization, admission control, and request routing.

kube-scheduler : Assigns pods to suitable nodes based on resource availability and policies.

kube-controller-manager : Maintains desired state, performs health checks, auto‑scaling, and rolling updates.

etcd : Distributed key‑value store persisting cluster state; requires regular backups and is used by control‑plane components and network plugins.

Worker Node

kubelet : Communicates with the API server to create and manage pods on the node.

kube-proxy : Implements Service networking, translating Service definitions into iptables/ipvs rules for load balancing.

Container runtime (Docker or other CRI): Executes containers.

Add‑on Components

CoreDNS : DNS service for cluster name resolution.

Kubernetes Dashboard : Web UI for resource management.

Ingress Controller : Provides Layer‑7 load balancing (e.g., NGINX, Traefik).

Monitoring stack : Prometheus, Alertmanager, Grafana.

Logging stack : EFK (Elasticsearch, Fluentd, Kibana).