Scaling

This article explains how to scale applications in OpenShift and Kubernetes. A Deployment is the common higher-level controller used to manage application replicas. A Deployment creates and manages a ReplicaSet (and previously a ReplicationController) which ensures the desired number of Pod replicas exist. To scale an application, you change the desired replica count in the controller’s spec; the ReplicaSet will then create or remove Pods to match that target. Key concepts:

Deployment: declarative controller that manages ReplicaSets and rolling updates.
ReplicaSet: ensures a specific number of Pod replicas run.
Service / Router: distributes traffic to Pods (internal via Service, external via OpenShift Router).

You can change the replica count in several common ways:

Edit the Deployment YAML and update spec.replicas.
Use command-line shortcuts such as oc scale or kubectl scale.
Use the OpenShift web console for one-click scaling.
Configure automatic scaling with a Horizontal Pod Autoscaler (HPA) to scale based on CPU, memory, or custom metrics.

A slide titled "Scale Deployment" showing a diagram of a deployment/replication controller with stacked pod/container icons on the left and, on the right, a webapp dashboard for a "simple-webapp" deployment displaying details and a circle indicating 5 pods.

Practical examples

Example Deployment YAML (relevant parts shown):

apiVersion: apps/v1
kind: Deployment
metadata:
  name: simple-webapp
spec:
  replicas: 3
  selector:
    matchLabels:
      app: simple-webapp
  template:
    metadata:
      labels:
        app: simple-webapp
    spec:
      containers:
      - name: web
        image: nginx:1.21
        ports:
        - containerPort: 80

Scale from the command line:

# Scale to 5 replicas (OpenShift)
oc scale deployment/simple-webapp --replicas=5

# Or with kubectl
kubectl scale deployment/simple-webapp --replicas=5

Query the current state:

# Check the Deployment and its replica count
oc get deployment simple-webapp

# List Pods created by the ReplicaSet
oc get pods -l app=simple-webapp

Enable automatic scaling with a Horizontal Pod Autoscaler (HPA):

# Create an HPA that scales between 2 and 10 replicas targeting 80% CPU
oc autoscale deployment/simple-webapp --min=2 --max=10 --cpu-percent=80

Summary table — scaling methods

Method	When to use	Example / Command
Manual edit (YAML)	Declarative change checked into Git / CI	Edit `spec.replicas` in the Deployment manifest
CLI scale	Quick, ad-hoc changes or scripts	`oc scale deployment/simple-webapp --replicas=5`
Web console	One-off interactive scaling	Use the OpenShift console’s scale control
HPA (autoscaling)	Scale based on CPU/memory/custom metrics	`oc autoscale deployment/simple-webapp --min=2 --max=10 --cpu-percent=80`

Links and references

Using the OpenShift web console is the quickest way for interactive scaling. For reproducible automation and CI/CD, prefer oc/kubectl commands or defining an HPA in YAML and managing it via GitOps.

Before scaling up, verify your project/cluster has sufficient resource quotas (CPU, memory) and available nodes. When scaling down, be careful with stateful workloads—Pods may be terminated. Use PodDisruptionBudgets and safe shutdown procedures for stateful applications.

Now let’s take a look at a demo.

Introduction

Pre-Requisite – Docker and Kubernetes

Getting Started with OpenShift

OpenShift Concepts – Projects and Users

Concepts – Builds and Deployments

Networks, Services, Routes and Scaling

Storage, Templates and Catalog

OpenShift Security

Conclusion and APPENDIX

Watch Video