Brief Overview on Kubernetes

Kubernetes is an open-source container orchestration platform initially developed by Google and now maintained by the Cloud Native Computing Foundation (CNCF). It automates deployment, scaling, and management of containerized applications, enabling resilient, portable workloads across on-premises and cloud environments.

Kubernetes Architecture and Cluster Components

A Kubernetes cluster consists of control plane nodes and worker nodes. Control plane components manage and maintain cluster state, while worker nodes run application containers.

Component	Role	Notes
kube-apiserver	Central API endpoint	Validates and configures data for API objects.
etcd	Distributed key–value store	Stores all cluster data and configuration.
kube-controller-manager	Control loops to enforce desired state	Manages node health, replication, and endpoint tracking.
kube-scheduler	Pod placement based on resource availability	Assigns pods to suitable nodes.
kubelet (worker)	Node agent ensuring containers run as expected	Registers node with control plane and reports status.
kube-proxy (worker)	Networking and load balancing for pods	Maintains network rules on nodes.
Container runtime (worker)	Executes container images (e.g., Docker, CRI-O)	Interfaces with kubelet to pull and run containers.

The image is a diagram illustrating the basics of Kubernetes architecture, showing the interaction between developers, admins, and operations with the controller node and worker nodes, including components like etcd, kube apiserver, and pods.

Pods and Workloads

A Pod is the smallest deployable unit in Kubernetes, encapsulating one or more containers that share:

A network namespace (IP address and ports)
Shared storage volumes
Container execution configuration

Pods are ephemeral by default; they do not self-heal once terminated.

Note

To achieve high availability and self-healing, wrap pods in higher-level controllers such as ReplicaSets and Deployments.

ReplicaSets and Deployments

ReplicaSet: Ensures a specified number of pod replicas are running at any time.
Deployment: Declaratively manages ReplicaSets to facilitate rolling updates and rollbacks.

Example Deployment manifest:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  replicas: 3
  selector:
    matchLabels:
      app: nginx
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:latest
        ports:
        - containerPort: 80

Service Discovery and Networking

Kubernetes Services provide stable endpoints to access pods. Common types include:

Service Type	Scope	Use Case
ClusterIP	Internal cluster only	Microservices communicating within the cluster.
NodePort	Static port on every node	Development or simple external access.
LoadBalancer	External cloud load balancer	Production traffic with a single public IP.

Warning

Provisioning a LoadBalancer service may incur additional cloud provider costs. Review your cloud network pricing before use.

Ingress

Ingress resources enable advanced HTTP(S) routing:

Path-based routing: Route requests by URL path (e.g., /app1, /app2).
Host-based routing: Direct traffic based on hostname (e.g., app.example.com).
TLS termination: Consolidate HTTPS certificates at the edge.

Ingress controllers often pair with ClusterIP services to secure internal access and reduce the number of external load balancers.

Links and References

Watch Video

Watch video content