Azure Kubernetes Service

Modern applications are increasingly built as microservices — many small, independent services rather than one large monolithic application. Running, scaling, and managing dozens or hundreds of containers across multiple servers by hand is impractical. Kubernetes is the industry-standard container orchestration platform that automates this. Azure Kubernetes Service (AKS) is Microsoft's managed Kubernetes offering — delivering the power of Kubernetes without the complexity of managing the control plane.

Understanding Containers First

A container packages an application and all its dependencies (libraries, runtime, configuration) into a single lightweight, portable unit. Containers run identically on any machine — a developer's laptop, a test server, or a cloud data center. Docker is the most common container runtime.

Virtual Machines vs Containers

  Virtual Machine:                Container:
  ┌─────────────────────┐        ┌─────────────────────┐
  │  App A              │        │  App A   App B      │
  │  ─────────          │        │  ───────  ───────   │
  │  OS (Full Windows   │        │  Libs A   Libs B    │
  │  or Linux — 20GB)   │        │  ─────────────────  │
  └─────────────────────┘        │  Container Runtime  │
  ┌─────────────────────┐        │  (Docker)           │
  │  App B              │        │  ─────────────────  │
  │  OS (Full — 20GB)   │        │  Host OS (shared)   │
  └─────────────────────┘        └─────────────────────┘
  Each VM has its own OS.        Containers share the host OS.
  Heavy, slow to start (~mins)   Lightweight, fast (~seconds)

What is Kubernetes?

Kubernetes (also called K8s) is an open-source platform for managing containerized applications at scale. It handles:

• Scheduling containers onto the right servers
• Restarting containers if they crash
• Scaling the number of running containers up or down
• Load balancing traffic across container instances
• Rolling out new versions without downtime
• Managing secrets and configuration for containers

AKS Architecture

  AKS Cluster
  │
  ├── Control Plane (Managed by Microsoft — FREE)
  │   ├── API Server     ← Receives all kubectl commands
  │   ├── Scheduler      ← Decides which node to run a pod on
  │   ├── etcd           ← Stores cluster state (key-value store)
  │   └── Controller Mgr ← Watches for desired state and corrects drift
  │
  └── Node Pool (Worker Nodes — You pay for these VMs)
      ├── Node 1 (VM: Standard_D4s_v5)
      │   ├── Pod: web-app (Container: nginx)
      │   └── Pod: api-service (Container: node:18)
      ├── Node 2 (VM: Standard_D4s_v5)
      │   ├── Pod: web-app (Container: nginx)
      │   └── Pod: database-worker (Container: python)
      └── Node 3 (VM: Standard_D4s_v5)
          └── Pod: api-service (Container: node:18)

Key Kubernetes Concepts

Creating an AKS Cluster

  # Create AKS cluster with 3 nodes
  az aks create \
    --resource-group myRG \
    --name myAKSCluster \
    --node-count 3 \
    --node-vm-size Standard_D2s_v5 \
    --enable-addons monitoring \
    --generate-ssh-keys

  # Download credentials to connect with kubectl
  az aks get-credentials \
    --resource-group myRG \
    --name myAKSCluster

  # Verify connection — list all nodes
  kubectl get nodes
  # NAME                     STATUS   ROLES   AGE
  # aks-nodepool1-12345-0    Ready    agent   2m
  # aks-nodepool1-12345-1    Ready    agent   2m
  # aks-nodepool1-12345-2    Ready    agent   2m

Deploying an Application to AKS

  # deployment.yaml
  apiVersion: apps/v1
  kind: Deployment
  metadata:
    name: web-app
  spec:
    replicas: 3          # Run 3 identical pods
    selector:
      matchLabels:
        app: web-app
    template:
      metadata:
        labels:
          app: web-app
      spec:
        containers:
        - name: web-app
          image: myacr.azurecr.io/webapp:v1.0
          ports:
          - containerPort: 80
  ---
  apiVersion: v1
  kind: Service
  metadata:
    name: web-app-service
  spec:
    type: LoadBalancer      # Creates a public Azure Load Balancer
    selector:
      app: web-app
    ports:
    - port: 80
      targetPort: 80

  # Apply the deployment
  kubectl apply -f deployment.yaml

  # Check status
  kubectl get pods
  kubectl get service web-app-service   # Shows the public IP after a minute

AKS Autoscaling

AKS supports two levels of autoscaling:

• Horizontal Pod Autoscaler (HPA): Automatically increases or decreases the number of pod replicas based on CPU or memory usage. When traffic spikes, more pods are created. When traffic drops, pods are removed.
• Cluster Autoscaler: Automatically adds or removes nodes (VMs) from the node pool. If pods cannot be scheduled because all nodes are full, a new node is added. If nodes are underutilized, they are removed to save cost.

AKS Integration with Azure Services

Key Takeaways

• AKS is a managed Kubernetes service — Microsoft manages the control plane for free; the user pays only for worker nodes.
• Kubernetes manages container scheduling, scaling, self-healing, and rolling deployments automatically.
• Key concepts: Pod (container group), Deployment (manages replicas), Service (stable network endpoint), Namespace (logical isolation).
• Horizontal Pod Autoscaler scales pods based on CPU/memory; Cluster Autoscaler scales the number of nodes.
• AKS integrates natively with Azure Container Registry, Azure AD, Key Vault, Monitor, and Virtual Networks.