Installing IBM Cloud Pak for Data and watsonx.data on Red Hat OpenShift

A cluster administrator can install IBM® watsonx.data on a Red Hat OpenShift cluster.

watsonx.data on Red Hat® OpenShift®

Before you begin

Install the IBM Cloud Pak for Data command-line interface.
Set up the installation environment variables.
Determine which watsonx.data components to install.
To install watsonx.data on Red Hat OpenShift Kubernetes Service (ROKS), ensure to meet the prerequisite.
Draft comment: arun.c.r@ibm.com
https://github.ibm.com/lakehouse/tracker/issues/7603

Restriction: If you are installing Red Hat OpenShift Container Platform Version 4.14, do not install the KubeVirt HyperConverged Cluster Operator on the cluster. It can cause problems when installing some Cloud Pak for Data software. (A previous version of this note stated that you could not install the OpenShift Virtualization Operator. However, that restriction no longer applies.)

Important: IBM watsonx.data can be co-located with watsonx.ai and watsonx.governance, and any supporting services that are included in the watsonx.data license. IBM watsonx.data cannot be co-located with Cloud Pak for Data Enterprise Edition, Cloud Pak for Data Standard Edition, or services included in other cartridge licenses. If you have an existing Cloud Pak for Data Enterprise Edition or Cloud Pak for Data Standard Edition installation, you must install the control plane and watsonx.data in its own operators project and operands project (namespace). Services included in the IBM watsonx.data license: Analytics Engine powered by Apache Spark (Starting with version 1.1.4, Analytics Engine powered by Apache Spark is automatically installed when you install watsonx.data.)

Note: If you are installing services other than watsonx.data, Analytics Engine, watsonx.ai, or watsonx.ai dependencies, install those in their own namespaces and then set $PROJECT_CPD_INST_OPERATORS and $PROJECT_CPD_INST_OPERANDS to different namespaces and run the installation steps.

Procedure

Log in to the ocp cluster as a user with sufficient permissions to complete this task.
Option 1: Run the following command to log in to the cluster by providing a username and password:
```
cpd-cli manage login-to-ocp \
--username=${OCP_USERNAME} \
--password=${OCP_PASSWORD} \
--server=${OCP_URL}
```
Option 2: Run the following command to log in to the cluster by providing a token:
```
cpd-cli manage login-to-ocp \
--server=${OCP_URL} \
--token=${OCP_TOKEN}
```

Install the certificate manager and the license service.

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

Optional: Install the scheduling service.

cpd-cli manage apply-scheduler \
--release=${VERSION} \
--license_acceptance=true \
--scheduler_ns=${PROJECT_SCHEDULING_SERVICE}

Run the cpd-cli manage authorize-instance-topology command to apply the required permissions to the projects.

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

Run the cpd-cli manage setup-instance-topology command to install IBM Cloud Pak foundational services and create the ConfigMap.

cpd-cli manage setup-instance-topology \
--release=${VERSION} \
--cpd_operator_ns=${PROJECT_CPD_INST_OPERATORS} \
--cpd_instance_ns=${PROJECT_CPD_INST_OPERANDS} \
--license_acceptance=true \
--block_storage_class=${STG_CLASS_BLOCK}

To install Cloud Pak for Data control plane and watsonx.data, run the following command:
```
export COMPONENTS=cpd_platform,watsonx_data
```
Note:
- For version 1.1.3 and earlier, if Spark is required, add analyticsengine to the components list.
- For version 1.1.4 and later, Spark is installed automatically as a dependency of watsonx.data. Adding the analyticsengine component is not required.

Install Cloud Pak for Data platform operator and service operator.

cpd-cli manage apply-olm \
--release=${VERSION} \
--cpd_operator_ns=${PROJECT_CPD_INST_OPERATORS} \
--components=${COMPONENTS}

Install the operands in the operands project for the instance.

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

Apply the entitlement for watsonx.data on Red Hat OpenShift.

cpd-cli manage apply-entitlement \
--cpd_instance_ns=<project-name> \
[--entitlement=watsonx-data] \
[--production=true|false] \
[--preview=true|false]