Getting started with IBM Cloud Kubernetes Service
IBM Cloud Kubernetes Service is a managed Kubernetes service to create your own cluster of compute hosts where you can deploy and manage containerized apps on IBM Cloud. Combined with an intuitive user experience, built-in security and isolation, and advanced tools to secure, manage, and monitor your cluster workloads, you can rapidly deliver highly available and secure containerized apps in the public cloud.
Ready to create your first cluster?
If you already have an IBM Cloud account and want to get started immediately:
Don't have an account? Sign up for IBM Cloud
Use this page to move through the main setup flow: review the service, prepare your account, plan your environment, create a cluster, and deploy a sample app.
Review the basics
Start with the core concepts, terminology, and benefits. For more information, see Understanding IBM Cloud Kubernetes Service.
If you already know the basics, continue to the next section.
Prepare your account
Make sure that your IBM Cloud account is ready for cluster creation. For setup steps, see Preparing your account to create clusters.
You must be logged in to IBM Cloud before you create a cluster. If you don't have an account yet, you can create one.
Pricing considerations
Pricing varies based on worker node flavor, number of nodes, and infrastructure type. For more information, see Kubernetes Service pricing.
Choose your infrastructure type
Before you create a cluster, choose the infrastructure type that fits your environment.
- VPC
- Worker nodes are provisioned as virtual server instances on VPC infrastructure. VPC is the recommended infrastructure type for new clusters. It provides network isolation by default and supports the latest IBM Cloud Kubernetes Service features, including dedicated hosts and secure by default.
- Classic
- Worker nodes are provisioned on virtual or bare metal machines on IBM Cloud classic infrastructure. Use Classic if you require bare metal worker nodes, software-defined storage (SDS) flavors, or connections to existing classic infrastructure resources.
- Satellite
- Worker nodes are provisioned on your own infrastructure, including on-premises hardware or virtual machines in other cloud providers such as AWS, Azure, and GCP. Use Satellite when you need to run IBM Cloud-managed clusters outside of IBM Cloud data centers.
For a detailed comparison including worker node options and availability, see Cluster environment strategy.
Create a cluster
Review your deployment options in the following table.
| Type | Level | Time | Description |
|---|---|---|---|
| Tutorial | Beginner | 1 hour | Follow the steps in this tutorial to create your own Virtual Private Cloud (VPC), then create an IBM Cloud Kubernetes Service cluster by using the CLI. For more information, see Create a cluster in your own Virtual Private Cloud. |
| Custom deployment | Intermediate | 1-3 hours | Create a custom cluster on Classic infrastructure. |
| Custom deployment | Intermediate | 1-3 hours | Create a custom cluster on VPC infrastructure. |
Already have a cluster? Learn how to access it and continue to the next step to deploy a sample app.
Deploy a sample app
After you create a cluster, deploy a sample app by using the Kubernetes dashboard.
- Select your cluster from the cluster list.
- Click Kubernetes dashboard.
- Click the Create new resource icon (
+), then select the Create from form tab. - Enter a name for your app, such as
liberty. - Enter
websphere-libertyfor the container image. - Enter the number of pods for your app deployment, such as
1. - From the Service list, select External to create a
LoadBalancerservice. - Configure the external service:
- Port:
80 - Target port:
9080 - Protocol:
TCP
- Port:
- Click Deploy.
- From the Pods menu, confirm that your
libertypod status is Running. - From the Services menu, open the External Endpoint of your
libertyservice. For example,169.xx.xxx.xxx:80for classic clusters orhttp://<hash>-<region>.lb.appdomain.cloud/for VPC clusters.
Your app is exposed by a Layer 4, version 1.0 network load balancer. To learn about other exposure options, see Planning in-cluster and external networking for apps.
Quick actions
If you already have a cluster, use these links to continue:
- Access your cluster
- Connect to your cluster and run kubectl commands. For more information, see Accessing clusters.
- Install the CLI
- Set up your local development environment. For more information, see Installing the CLI.
- Deploy your first app
- Get hands-on with a sample application. For more information, see Deploying apps.
Need to check pricing first? View Kubernetes Service pricing.
What's next?
Continue with one of these curated learning paths: