Cluster Autoscaler concept in Kubernetes - Mini Project: Build & Apply

Cluster Autoscaler Concept in Kubernetes

📖 Scenario: You manage a Kubernetes cluster that runs several applications. Sometimes, the cluster needs more nodes to handle extra work, and sometimes it has extra nodes that are not needed. To save resources and keep the cluster efficient, you want to use the Cluster Autoscaler.The Cluster Autoscaler automatically adjusts the number of nodes in your cluster based on the workload. It adds nodes when pods cannot be scheduled due to lack of resources and removes nodes when they are underutilized.

🎯 Goal: In this project, you will create a simple Kubernetes deployment and configure a Cluster Autoscaler setting to understand how it works. You will simulate the autoscaler behavior by defining node group size limits and observe how the autoscaler would react.

📋 What You'll Learn

Create a Kubernetes deployment with a fixed number of replicas

Define a node group size configuration with minimum and maximum nodes

Write a simple logic to simulate scaling decisions based on pod resource requests

Output the scaling decision result

💡 Why This Matters

🌍 Real World

Cluster Autoscaler helps keep Kubernetes clusters efficient by automatically adjusting node counts based on workload demands.

💼 Career

Understanding Cluster Autoscaler is essential for DevOps engineers managing cloud-native applications to optimize resource usage and cost.

1

Create a Kubernetes deployment YAML

Create a variable called deployment_yaml that contains a string with a Kubernetes deployment manifest. The deployment should be named web-app, have 3 replicas, and use the image nginx:latest.

Kubernetes

# Your code here

Need a hint?

Use triple quotes to create a multi-line string for the YAML manifest.

2

Define node group size limits

Create two variables called min_nodes and max_nodes and set them to 1 and 5 respectively. These represent the minimum and maximum number of nodes the autoscaler can manage.

Kubernetes

deployment_yaml = """
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-app
spec:
  replicas: 3
  selector:
    matchLabels:
      app: web-app
  template:
    metadata:
      labels:
        app: web-app
    spec:
      containers:
      - name: nginx
        image: nginx:latest
"""
# Define min_nodes and max_nodes variables
# Your code here

Need a hint?

Just assign the numbers 1 and 5 to the variables min_nodes and max_nodes.

3

Simulate autoscaler decision logic

Create a variable called current_nodes set to 1. Then write an if statement that checks if the number of replicas in the deployment (3) is greater than current_nodes * 2. If true, set a variable scale_up to True, else set it to False.

Kubernetes

deployment_yaml = """
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-app
spec:
  replicas: 3
  selector:
    matchLabels:
      app: web-app
  template:
    metadata:
      labels:
        app: web-app
    spec:
      containers:
      - name: nginx
        image: nginx:latest
"""

min_nodes = 1
max_nodes = 5

# Set current_nodes and write if statement to set scale_up
# Your code here

Need a hint?

Use the number 3 directly for replicas since it is fixed in the deployment.

4

Output the autoscaler decision

Write a print statement that outputs "Scale up needed" if scale_up is True, otherwise output "No scaling needed".

Kubernetes

deployment_yaml = """
apiVersion: apps/v1
kind: Deployment
metadata:
  name: web-app
spec:
  replicas: 3
  selector:
    matchLabels:
      app: web-app
  template:
    metadata:
      labels:
        app: web-app
    spec:
      containers:
      - name: nginx
        image: nginx:latest
"""

min_nodes = 1
max_nodes = 5

current_nodes = 1
if 3 > current_nodes * 2:
    scale_up = True
else:
    scale_up = False

# Print the scaling decision
# Your code here

Need a hint?

Use an if statement to print the correct message based on scale_up.