Tutorial

Install IBM Cloud Pak for Data 5.3 on Red Hat OpenShift Container Platform 4.20 on IBM Power Virtual Server

Bringing innovations for data and AI to the cloud

By Claus Huempel, Daniel Casali

Introduction

IBM Cloud Pak for Data unifies and simplifies the collection, organization, and analysis of data. Enterprises can turn data into insights through an integrated cloud-native architecture. IBM Cloud Pak for Data is extensible and can be customized to a client's unique data and AI landscapes through an integrated catalogue of IBM, open source, and third-party microservices add-ons.

This tutorial shows how to perform an online installation of IBM Cloud Pak for Data 5.3 on Red Hat OpenShift Container Platform 4.20 running on IBM Power Virtual Server and some of the services that are needed to use the IBM Industry Accelerators.

Prerequisites

This tutorial assumes that you are familiar with the Red Hat OpenShift Container Platform 4.20 environment on IBM Power Virtual Server. It is assumed that you have it already installed, you have access to it, and have the credentials of an OpenShift cluster administrator (kubeadmin). You must be familiar with Linux command line and have at least a basic understanding of Red Hat OpenShift.

For this tutorial, we assume that the OpenShift 4.20 environment on IBM Power Virtual Server consists of three worker nodes, each with two physical IBM Power cores (that is, 16 vCPU from a Kubernetes level) and 64 GB RAM.

Also, you must have created a local repository on a persistent storage and have a Network File System (NFS) storage class where the NFS export has the no_root_squash property set.

You need to have the wget and the oc clients already installed on your PATH variable.

Estimated time

It is expected to take around 2 to 3 hours to complete the installation of IBM Cloud Pak for Data 5.3 on IBM Power Virtual Server. This lengthy duration is because we need to install the software from internet repositories.

Steps

Installation of IBM Cloud Pak for Data 5.3 on IBM Power Virtual Server includes the following steps:

Prepare the bastion host.
Prepare the OpenShift cluster.
Install IBM Software Hub (formerly known as IBM Cloud Pak for Data Platform).
Install IBM Watson Studio and IBM Watson Machine Learning services.
Run an AutoAI experiment.
Optional: Install additional RStudio, Analytics Engine, Decision Optimization, Orchestration Pipelines, IBM Db2 services, DataStage and IBM Knowledge Catalog.

Step 1 – Prepare the bastion host

Install the Linux screen utility that will maintain your session in case your internet connection drops and will make it recoverable when you reconnect.

# on a RHEL 8 system run:
yum -y install https://dl.fedoraproject.org/pub/epel/8/Everything/ppc64le/Packages/s/screen-4.6.2-12.el8.ppc64le.rpm
# on a RHEL 9 system run:
yum -y install https://dl.fedoraproject.org/pub/epel/9/Everything/ppc64le/Packages/s/screen-4.8.0-6.el9.ppc64le.rpm

Install the Linux podman package that will allow you to run containers on the bastion host. The Cloud Pak for Data CLI software, that we will be using to install the IBM Cloud Pak for Data software later in this tutorial, runs inside a container on the bastion host.
yum -y install podman

Verify that the no_root_squash parameter is set for the NFS share. Change the parameter in the file /etc/exports to no_root_squash if needed.

sed -i 's/,root_squash/,no_root_squash/g' /etc/exports
exportfs -a
exportfs -v

Create a new user cp4d on the bastion host, that we will be further using in the installation process, and change to that new user.
useradd cp4d
Create a password for the new cp4d user.
passwd cp4d
Log in to the cp4d user using an SSH session (we need to use an SSH session here as we are going to run non-root podman commands using the cpd-cli command in chapter 2 of this tutorial and changing to the new user using the su – cp4d command may result in non-root podman commands not working correctly).
ssh cp4d@localhost

Download the cpd-cli utility from GitHub:

wget https://github.com/IBM/cpd-cli/releases/download/v14.3.0_refresh_2/cpd-cli-ppc64le-EE-14.3.0.tgz
tar -xzvf cpd-cli-ppc64le-EE-14.3.0.tgz
mv cpd-cli-ppc64le-EE-14.3.0*/* .
rm -rf cpd-cli-ppc64le-EE-14.3.0*
rm -f cpd-cli-ppc64le-EE-14.3.0.tgz
./cpd-cli version

After running the commands, you should get an output like to the following sample:

cpd-cli
     Version: 14.3.0
     Build Date: 2025-12-10T14:05:48
     Build Number: 2819
     SWH Release Version: 5.3.0

Create a file, pvs_vars.sh, and open it with your favourite editor.

Retrieve the OpenShift Container Platform API url. Log in to the OpenShift web console as kubeadmin user. Then click the Copy login command under the “kube:admin” widget at the upper-right corner. Click Display token. Under “Log in with this token”, you should see an entry similar to this:

oc login --token=sha256~kOnz77iElG4JrlFxEccBTmi_pDbw9Q8IeGJc_S5urfc --server=https://api.itzpvs-218018.cecc.ihost.com:6443

Make a note of the OpenShift Container Platform API url value (the value after the --server= parameter including the port number) and store it in the PVC_API_url environment variable in your pvs_vars.sh file:

export PVS_API_url=<replace with the value of the OCP API url>
# for example:
export PVS_API_url=https://api.itzpvs-218018.cecc.ihost.com:6443

Retrieve the kubeadmin password and store it in the PVS_CLUSTER_ADMIN_PWD environment variable in your pvs_vars.sh file:

export PVS_CLUSTER_ADMIN_PWD=<replace with your cluster admin password>
# for example:
export PVS_CLUSTER_ADMIN_PWD=wwwww-xxxxx-yyyyy-zzzzz

Obtain your IBM entitlement API key for the IBM Cloud Pak for Data software and store the value in the PVS_IBM_ENTITLEMENT_KEY environment variable your pvs_vars.sh file. You can find the instructions for obtaining the key at: https://www.ibm.com/docs/en/software-hub/5.3.x?topic=information-obtaining-your-entitlement-api-key.

export PVS_IBM_ENTITLEMENT_KEY=<replace with the value of your IBM entitlement API key>
# for example:
export PVS_IBM_ENTITLEMENT_KEY=eyJ0eXAiOiJKV1QiLCJxxx

Define the PVS_API_HOST environment variable and store it in your pvs_vars.sh file:

export PVS_API_HOST=<replace with the hostname of the API server>

# for example:
export PVS_API_HOST=api.itzpvs-218018.cecc.ihost.com

Finally, define the PVS_ENCODED_IBM_ENTITLEMENT_KEY environment variable by copying and pasting the following line into your pvs_vars.sh file:

export PVS_ENCODED_IBM_ENTITLEMENT_KEY=$(echo -n "cp:$PVS_IBM_ENTITLEMENT_KEY" | base64 -w 0)

Save your changes to the pvs_vars.sh file.
Source the pvs_vars.sh file.
source pvs_vars.sh

Verify all five environment variables.

set | grep PVS

# you should see an output like the following:
PVS_API_HOST=api.itzpvs-218018.cecc.ihost.com
PVS_API_url=https://api.itzpvs-218018.cecc.ihost.com:6443
PVS_CLUSTER_ADMIN_PWD=qD8nz-aDQxj-rxeVB-D8S3f
PVS_ENCODED_IBM_ENTITLEMENT_KEY=Y3A6ZXlKMGVYxxxxxxxx
PVS_IBM_ENTITLEMENT_KEY=eyJ0eXAiOiJKV1QiLCJxxxxxxxxx

Validate that you can successfully log in to your OpenShift cluster using the oc command.

oc login -u kubeadmin -p $PVS_CLUSTER_ADMIN_PWD $PVS_API_url --insecure-skip-tls-verify

# you should see an output like the following:
Login successful.

You have access to 75 projects, the list has been suppressed. You can list all projects with 'oc projects'

Using project "default".
Welcome! See 'oc help' to get started

Verify that the default storage class is nfs-storage-provisioner.

oc get sc

# you should see an output like the following:
NAME                                PROVISIONER   RECLAIMPOLICY   VOLUMEBINDINGMODE   ALLOWVOLUMEEXPANSION   AGE
nfs-storage-provisioner (default)   nfs-storage   Delete          Immediate           false                  2d1h

Generate a cpd_vars.sh environment variables file by running the following command:

tee cpd_vars.sh <<EOF
# ==========================================================
# IBM Software Hub installation variables
# ==========================================================

# ----------------------------------------------------------
# Client workstation
# ----------------------------------------------------------
export OLM_UTILS_LAUNCH_ARGS="--network host"

# ----------------------------------------------------------
# Cluster
# ----------------------------------------------------------
export OCP_url="$PVS_API_HOST:6443"
export OPENSHIFT_TYPE="self-managed"
export OCP_USERNAME="kubeadmin"
export OCP_PASSWORD="$PVS_CLUSTER_ADMIN_PWD"
export OCP_TOKEN="$(oc whoami -t)"

# ----------------------------------------------------------
# Projects
# ----------------------------------------------------------
export PROJECT_CERT_MANAGER="cert-manager"
export PROJECT_LICENSE_SERVICE="ibm-licensing"
export PROJECT_SCHEDULING_SERVICE="cpd-scheduler"
export PROJECT_CPD_INST_OPERATORS="cpd-operators"
export PROJECT_CPD_INST_OPERANDS="cpd-instance"

# ----------------------------------------------------------
# Storage
# ----------------------------------------------------------
export STG_CLASS_BLOCK=nfs-storage-provisioner
export STG_CLASS_FILE=nfs-storage-provisioner

# ----------------------------------------------------------
# IBM Entitled Registry
# ----------------------------------------------------------
export IBM_ENTITLEMENT_KEY=$PVS_IBM_ENTITLEMENT_KEY

# ----------------------------------------------------------
# Image pull configuration
# ----------------------------------------------------------
export IMAGE_PULL_SECRET=cpd-image-pull-secret
export IMAGE_PULL_CREDENTIALS=$PVS_ENCODED_IBM_ENTITLEMENT_KEY
export IMAGE_PULL_PREFIX=icr.io

# ----------------------------------------------------------
# IBM Software Hub version
# ----------------------------------------------------------
export VERSION=5.3.0

# ----------------------------------------------------------
# Components
# ----------------------------------------------------------
export COMPONENTS=ws,wml
EOF

Verify your cpd_vars.sh file. The file should look similar to the following example:

cat cpd_vars.sh
# for example:
# ==========================================================
# IBM Software Hub installation variables
# ==========================================================

# ----------------------------------------------------------
# Client workstation
# ----------------------------------------------------------
export OLM_UTILS_LAUNCH_ARGS="--network host"

# ----------------------------------------------------------
# Cluster
# ----------------------------------------------------------
export OCP_url="api.itzpvs-218018.cecc.ihost.com:6443"
export OPENSHIFT_TYPE="self-managed"
export OCP_USERNAME="kubeadmin"
export OCP_PASSWORD="wwwww-xxxxx-yyyyy-zzzzz"
export OCP_TOKEN="sha256~xxxxxxxxxxxxxxxxxxx"

# ----------------------------------------------------------
# Projects
# ----------------------------------------------------------
export PROJECT_CERT_MANAGER="cert-manager"
export PROJECT_LICENSE_SERVICE="ibm-licensing"
export PROJECT_SCHEDULING_SERVICE="cpd-scheduler"
export PROJECT_CPD_INST_OPERATORS="cpd-operators"
export PROJECT_CPD_INST_OPERANDS="cpd-instance"

# ----------------------------------------------------------
# Storage
# ----------------------------------------------------------
export STG_CLASS_BLOCK=nfs-storage-provisioner
export STG_CLASS_FILE=nfs-storage-provisioner

# ----------------------------------------------------------
# IBM Entitled Registry
# ----------------------------------------------------------
export IBM_ENTITLEMENT_KEY=eyJ0eXAxxxxxxx

# ----------------------------------------------------------
# Image pull configuration
# ----------------------------------------------------------
export IMAGE_PULL_SECRET=cpd-image-pull-secret
export IMAGE_PULL_CREDENTIALS=Y3A6ZXlxxxxx
export IMAGE_PULL_PREFIX=icr.io

# ----------------------------------------------------------
# IBM Software Hub version
# ----------------------------------------------------------
export VERSION=5.3.0

# ----------------------------------------------------------
# Components
# ----------------------------------------------------------
export COMPONENTS=ws,wml

Step 2 – Prepare the OpenShift cluster

As cp4d user, source the cpd_vars.sh file.
source cpd_vars.sh

Run the cpd-cli manage login command. Note that running the command for the first time takes some time to complete, as the container image that is used by the cpd-cli manage login command will be downloaded from the internet.

./cpd-cli manage login-to-ocp --token=${OCP_TOKEN} --server=${OCP_url}

Successful completion of the cpd-cli manage login command would look like the following sample:

[…]
Using project "default" on server "https://api.itzpvs-218018.cecc.ihost.com:6443".
[SUCCESS] 2026-01-29T07:12:58.761817Z You may find output and logs in the /home/cp4d/cpd-cli-workspace/olm-utils-workspace/work directory.
[SUCCESS] 2026-01-29T07:12:58.761880Z The login-to-ocp command ran successfully.[

Add your IBM entitlement key to the global pull secret of your OpenShift cluster.

./cpd-cli manage add-icr-cred-to-global-pull-secret --entitled_registry_key=${IBM_ENTITLEMENT_KEY}

Successful completion would look as follows:

[SUCCESS] 2026-01-29T07:39:58.525721Z The add-icr-cred-to-global-pull-secret command ran successfully.

Create new OpenShift projects for the IBM Cloud Pak for Data installation.

oc new-project ${PROJECT_LICENSE_SERVICE}
oc new-project ${PROJECT_SCHEDULING_SERVICE}
oc new-project ${PROJECT_CPD_INST_OPERATORS}
oc new-project ${PROJECT_CPD_INST_OPERANDS}

Successful completion would look as follows:

[…]
Now using project "cpd-instance" on server "https://api.itzpvs-218018.cecc.ihost.com:6443".
[…]

Install the OpenShift Certificate Manager.

The installation of the OpenShift Certificate Manager consists of the following steps:

a. Create a new project for the OpenShift Certificate Manager operator.

oc new-project cert-manager-operator

b. Create a YAML file operatorGroup.yaml that defines an operator group for the OpenShift Certificate Manager operator.

tee operatorGroup.yaml <<EOF
apiVersion: operators.coreos.com/v1
kind: OperatorGroup
metadata:
  name: openshift-cert-manager-operator
  namespace: cert-manager-operator
spec:
  targetNamespaces: []
  spec: {}
EOF

c. Apply the operatorGroup.yaml YAML file. This will create the operator group.

oc create -f operatorGroup.yaml

Successful completion would look like:

operatorgroup.operators.coreos.com/openshift-cert-manager-operator created

d. Create a YAML file, named subscription.yaml, that defines a subscription to the OpenShift Certificate Manager operator.

tee subscription.yaml <<EOF
apiVersion: operators.coreos.com/v1alpha1
kind: Subscription
metadata:
  name: openshift-cert-manager-operator
  namespace: cert-manager-operator
spec:
  channel: stable-v1
  name: openshift-cert-manager-operator
  source: redhat-operators
  sourceNamespace: openshift-marketplace
  installPlanApproval: Automatic
EOF

e. Apply the subscription.yaml YAML file. This will create the subscription and install the OpenShift Certificate Manager including its operator.

Successful completion would look as follows:

subscription.operators.coreos.com/openshift-cert-manager-operator created

f. Verify that the OpenShift Certificate Manager has been installed and is running. The pods of the OpenShift Certificate Manager might take 1-2 minutes to come up.

oc get pods -n cert-manager

Successful completion would look as follows:

Step 3 – Install IBM Software Hub and IBM Cloud Pak for Data control plane

As the cp4d user, open a screen session. This will allow you to reconnect to your terminal using the screen -r command if you lose the SSH connection to the bastion host.
screen

Install the shared cluster components. This will set up the Red Hat OpenShift certificate manager to be used and will install the IBM licensing service on your cluster.

./cpd-cli manage apply-cluster-components \
--release=${VERSION} \
--license_acceptance=true \
--cert_manager_ns=${PROJECT_CERT_MANAGER} \
--licensing_ns=${PROJECT_LICENSE_SERVICE}

Successful completion will look as follows:

[SUCCESS] 2026-01-29T09:55:53.204829Z The apply-cluster-components command ran successfully.

Check the health of your OpenShift cluster to make sure that everything is working properly before installing IBM Software Hub and the IBM Cloud Pak for Data control plane.
./cpd-cli health cluster
Successful completion will look as follows:

Authorize the instance topology.

./cpd-cli manage authorize-instance-topology \
--cpd_operator_ns=${PROJECT_CPD_INST_OPERATORS} \
--cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS}

Successful completion will look as follows:

[SUCCESS] 2026-01-29T10:07:14.788528Z The authorize-instance-topology command ran successfully.

Download the CASE packages.

./cpd-cli manage case-download \
--components=cpd_platform,${COMPONENTS} \
--release=${VERSION} \
--from_oci=true \
--operator_ns=${PROJECT_CPD_INST_OPERATORS} \
--cluster_resources=true

Successful completion will look like this:

[✔] /home/cp4d/cpd-cli-workspace/olm-utils-workspace/work/cluster_scoped_resources.yaml is generated. To apply the file to cluster:
oc apply -f /home/cp4d/cpd-cli-workspace/olm-utils-workspace/work/cluster_scoped_resources.yaml

[SUCCESS] 2026-01-29T10:15:59.787682Z You may find output and logs in the /home/cp4d/cpd-cli-workspace/olm-utils-workspace/work directory.
[SUCCESS] 2026-01-29T10:15:59.787736Z The case-download command ran successfully.

Apply the cluster-scoped resources. The prior case-download command creates a new YAML file, named cluster_scoped_resources.yaml, in the work directory. Run the following command to apply this YAML file to the cluster.

oc apply -f /home/cp4d/cpd-cli-workspace/olm-utils-workspace/work/cluster_scoped_resources.yaml \
--server-side \
--force-conflicts

Successful completion of the command would look as follows:

[…]
customresourcedefinition.apiextensions.k8s.io/ibmusagemeterings.operator.ibm.com created

Create a new JSON file, named dockerconfig.json.

tee dockerconfig.json <<EOF
{
  "auths": {
    "cp.icr.io": {
      "auth": "${IMAGE_PULL_CREDENTIALS}"
    },
    "icr.io":{
      "auth": "${IMAGE_PULL_CREDENTIALS}"
    }
  }
}
EOF

Create two new pull secrets.

oc create secret docker-registry ${IMAGE_PULL_SECRET} \
--from-file ".dockerconfigjson=dockerconfig.json" \
--namespace=${PROJECT_CPD_INST_OPERATORS}

oc create secret docker-registry ${IMAGE_PULL_SECRET} \
--from-file ".dockerconfigjson=dockerconfig.json" \
--namespace=${PROJECT_CPD_INST_OPERANDS}

Successful completion would look as follows:

secret/cpd-image-pull-secret created
secret/cpd-image-pull-secret created

Install the Cloud Pak for Data platform components and include running a storage validation test. Note that this step can take 30-60 minutes as container images will be downloaded from the IBM container registry and containerized software for the IBM Cloud Pak for Data instance will be deployed into the PROJECT_CPD_INST_OPERANDS namespace.

./cpd-cli manage install-components \
--license_acceptance=true \
--components=cpd_platform \
--release=${VERSION} \
--operator_ns=${PROJECT_CPD_INST_OPERATORS} \
--instance_ns=${PROJECT_CPD_INST_OPERANDS} \
--block_storage_class=${STG_CLASS_BLOCK} \
--file_storage_class=${STG_CLASS_FILE} \
--image_pull_prefix=${IMAGE_PULL_PREFIX} \
--image_pull_secret=${IMAGE_PULL_SECRET} \
--run_storage_tests=true

Successful completion will look as follows:

[SUCCESS] 2026-01-29T11:26:51.150036Z The install-components command ran successfully.

Step 4 – Install Watson Studio and Watson Machine Learning services

Install Watson Studio and Watson Machine Learning services by running the install-components command again, but use the COMPONENTS. Note, that this command eventually takes 30 to 60 minutes to complete due to the number of services that will be installed during Watson Studio and Watson Machine Learning installation.

./cpd-cli manage apply-cr \
--release=${VERSION} \
--license_acceptance=true \
--cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
--components=${COMPONENTS} \
--block_storage_class=${STG_CLASS_BLOCK} \
--file_storage_class=${STG_CLASS_FILE} \
-v

Successful completion will look as follows:

[SUCCESS] 2026-01-29T12:08:04.556602Z The install-components command ran successfully.

Watch the installation progress. You may want to monitor the installation progress by running the following command in a second command line terminal in your bastion host. Make sure to run the source cpd_vars.sh command before, so that the PROJECT_CPD_INST_OPERANDS environment variable is defined. The command shows the list of pods in the IBM Cloud Pak for Data instance namespace with the most recently created pods being shown first:
watch -n 10 "oc get pods -n ${PROJECT_CPD_INST_OPERANDS} \ --sort-by=.status.startTime | tac"
Output:

Get the url of the web client and the automatically created password for the admin user:

./cpd-cli manage get-cpd-instance-details \
--cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
--get_admin_initial_credentials=true

Output:

CPD url: cpd-cpd-instance.apps.itzpvs-218018.cecc.ihost.com
CPD Username: cpadmin
CPD Password: xxx

Log in to the IBM Cloud Pak for Data web console.

Open a browser and enter the IBM Cloud Pak for Data web console CPD url you retrieved from the previous step prepended with “https://”. Accept the warnings of a potential security risk. Then enter cpadmin as the username and the default IBM Cloud Pak for Data admin password retrieved from step 3. Click Log in to proceed.
Click the Switch Location icon at the upper-right corner.

View larger image
Click IBM Cloud Pak for Data.
Notice that the IBM Cloud Pak for Data web console is displayed.

View larger image

Step 5 – Run an AutoAI experiment

For a quick test, you can run an AutoAI experiment. Perform the following steps to conduct the test using the AutoAI experiment.

Click the main menu icon at the upper-left corner.
Click Projects --> All projects.
Click New project.

View larger image
Enter autoai-test as the project name and enter a description. Click Create.

View larger image
Click the Assets tab.

View larger image
Click New asset.

View larger image
Click Work with models and then click the Automate the building of AI solutions with your data tile.

View larger image
Enter a name for your AutoAI experiment and click Create.

View larger image
Open a new tab in your browser and download the Titanic example.
Save the downloaded the CSV file and drop it on the Add data sources page.

View larger image

View larger image
In the Configure details section, click No for “Create a time series analysis?”.
Select Survived as the prediction column for “What do you want to predict?”.
Click Run experiment.

View larger image
Watch the progress of the AutoAI experiment. You might need to wait at the first time a bit in Pending state as a new container image for the AutoAI experiment is going to be downloaded from the internet to your OpenShift Container Platform cluster.

View larger image

View larger image

View larger image
After the experiment is complete, notice that the state now shows “Experiment completed” indicating that you have successfully run the AutoAI experiment on your IBM Cloud Pak for Data 5.3 cluster running on OpenShift 4.20 on Power Virtual Server.

Step 6 – Optional: Install Analytics Engine, RStudio, Decision Optimization, Orchestration Pipelines, IBM Db2 services, DataStage, and IBM Knowledge Catalog

Modify the cpd_vars.sh file in order to specify the additional IBM Cloud Pak for Data services you want to install on the cluster.

cp cpd_vars.sh cpd_vars.sh.ORG

sed -i \
's/COMPONENTS=ws,wml/COMPONENTS=analyticsengine,rstudio,dods,ws_pipelines,db2oltp,db2wh,dmc,datastage_ent_plus,wkc/g' \
cpd_vars.sh

In our case, for IBM Cloud Pak for Data on Power Virtual Server, we are going to install these additional services: Analytics Engine, RStudio, Decision Optimization, Orchestration Pipelines, Db2 Services (Db2 OLTP, Db2 Warehouse, and Db2 Management Console), DataStage, and IBM Knowledge Catalog.

Source the modified cpd_vars.sh file.
source cpd_vars.sh

Verify that the COMPONENTS environment variable has been updated by running the following command:

set | grep COMPONENTS

You should see an output like this:

COMPONENTS=analyticsengine,rstudio,dods,ws_pipelines,db2oltp,db2wh,dmc,datastage_ent_plus,wkc

Re-run the cpd-cli manage case-download command to download the CASE packages for the new CPD components.

./cpd-cli manage case-download \
--components=${COMPONENTS} \
--release=${VERSION} \
--from_oci=true \
--operator_ns=${PROJECT_CPD_INST_OPERATORS} \
--cluster_resources=true

Successful completion will look as follows:

[SUCCESS] 2026-01-29T15:22:52.281534Z The case-download command ran successfully.

Create the cluster-scope resources for the new components.

oc apply -f /home/cp4d/cpd-cli-workspace/olm-utils-workspace/work/cluster_scoped_resources.yaml \
--server-side \
--force-conflicts

Re-run the cpd-cli manage install-components command to install the new components.

./cpd-cli manage install-components \
--license_acceptance=true \
--components=${COMPONENTS} \
--release=${VERSION} \
--operator_ns=${PROJECT_CPD_INST_OPERATORS} \
--instance_ns=${PROJECT_CPD_INST_OPERANDS} \
--block_storage_class=${STG_CLASS_BLOCK} \
--file_storage_class=${STG_CLASS_FILE} \
--image_pull_prefix=${IMAGE_PULL_PREFIX} \
--image_pull_secret=${IMAGE_PULL_SECRET}

Successful completion will look as follows:

[SUCCESS] 2025-06-17T06:45:22.843084Z The apply-cr command ran successfully.

Next, you need to create a new Db2 database instance on the IBM Cloud Pak for Data cluster to verify that the Db2 service on IBM Cloud Pak for Data is working fine.

Log in to the IBM Cloud Pak for Data web console.
Click the main menu icon and then click Data --> Databases.
Click Create a database.

View larger image
Click Db2 and then click Next.

View larger image
Scroll down and click Single location for all data. Then, click Next.

View larger image
Click Next.

View larger image
From the Input method list, select Generate a Kubernetes secret. Click Next.

View larger image
On the Storage page, make sure that the required settings are selected and click Next.

View larger image
On the Finalize page, validate the options you have set for the database and click Create.

View larger image
The new Db2 database is now being provisioned. Note that the provisioning process can take 10 -15 minutes to complete.

Summary

This tutorial helped you to install a comprehensive AI and machine learning environment using IBM Cloud Pak for Data 5.3 on your IBM Power Virtual Server environment and to run a simple AutoAI experiment.

Topics

Languages

Products

Open Source

Install IBM Cloud Pak for Data 5.3 on Red Hat OpenShift Container Platform 4.20 on IBM Power Virtual Server

Introduction

Prerequisites

Estimated time

Steps

Step 1 – Prepare the bastion host

Step 2 – Prepare the OpenShift cluster

Step 3 – Install IBM Software Hub and IBM Cloud Pak for Data control plane

Step 4 – Install Watson Studio and Watson Machine Learning services

Step 5 – Run an AutoAI experiment

Step 6 – Optional: Install Analytics Engine, RStudio, Decision Optimization, Orchestration Pipelines, IBM Db2 services, DataStage, and IBM Knowledge Catalog

Summary

Further readings