Software requirements

Before you install IBM Cloud Pak® for Data, review the software requirements for the control plane, the shared cluster components, and the services that you plan to install, and review the supported web browsers.

Cloud Pak for Data platform software requirements

You must have the following software to install Cloud Pak for Data:

Red Hat® OpenShift® Container Platform cluster

For entitlement information see Licenses and entitlements.

Cloud Pak for Data supports the following versions of Red Hat OpenShift Container Platform. (Cloud Pak for Data supports the same operating system requirements as Red Hat OpenShift Container Platform.)

Important: Cloud Pak for Data supports only the specified releases of Red Hat OpenShift Container Platform.

Different versions of Cloud Pak for Data support different versions of Red Hat OpenShift Container Platform.

Version Supported on Learn more Cluster sizing guidance
Version 4.12.0 or later fixes
  • 4.8.0
  • 4.8.1
  • 4.8.2
  • 4.8.3
  • 4.8.4
  • 4.8.5
 For details, see the Red Hat OpenShift Container Platform documentation: Refer to the Cloud Pak for Data platform hardware requirements as you configure your cluster.
Version 4.14.0 or later fixes
  • 4.8.1
  • 4.8.2
  • 4.8.3
  • 4.8.4
  • 4.8.5
 For details, see the Red Hat OpenShift Container Platform documentation:
Restriction: Do not install the OpenShift Virtualization Operator on the cluster. It can cause problems when installing some Cloud Pak for Data software.
 Refer to the Cloud Pak for Data platform hardware requirements as you configure your cluster.
Version 4.15.0 or later fixes
  • 4.8.5
 For details, see the Red Hat OpenShift Container Platform documentation:
Restriction: Do not install the OpenShift Virtualization Operator on the cluster. It can cause problems when installing some Cloud Pak for Data software.
 Refer to the Cloud Pak for Data platform hardware requirements as you configure your cluster.
Kubernetes Metrics Server
If you do not enable the default monitoring stack on your Red Hat OpenShift Container Platform cluster and you want to gather use metrics for your pods and nodes, install Kubernetes Metrics Server.
Important: If you do not enable the default monitoring stack or install Kubernetes Metrics Server, the platform monitoring features in Cloud Pak for Data will not work.

Shared cluster component software requirements

There are no additional software requirements for:
  • Scheduling service
  • Common core services

Service software requirements

Use this table to determine whether the service that you want to install depends on other software being available:
  • Some services require other software to be installed outside of Cloud Pak for Data (marked as external dependencies)
  • Some services require other Cloud Pak for Data services to be installed as prerequisites or to support specific functionality (marked as service dependencies)
  • Some services require other underlying components, which the service installs if needed (marked as component dependencies)
Service External dependencies Service dependencies Component dependencies
AI Factsheets None
  • 4.8.0, 4.8.1, or 4.8.2 only. To install this service, you must have one of the following services installed:
    • IBM® Knowledge Catalog
    • Watson™ Studio
  • 4.8.3 or later. The service does not have any service dependencies.
To implement a complete AI governance solution, you must install the following services:
  • OpenPages®
  • IBM Knowledge Catalog
  • Watson OpenScale
  • Watson Machine Learning
  • Watson Studio

If you install AI Factsheets with only Watson Studio, you can create AI use cases in only one catalog.
None
  • 4.8.0, 4.8.1, or 4.8.2 only, the service does not have any component dependencies.
  • 4.8.3 or later. The service automatically installs the following dependencies if they are not already installed:
    • Common core services (ccs)
Anaconda Repository for IBM Cloud Pak for Data For details, see the Anaconda installation requirements.
None
None
Analytics Engine powered by Apache Spark None
None
None
Cognos® Analytics To use Cognos Analytics, you must have:
None
The service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
Cognos Dashboards None
None
The service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
  • Elasticsearch (opencontent_elasticsearch)
  • Redis (ibm_redis_cp)
Data Privacy None
To install this service, you must have the following services already installed:
  • Analytics Engine powered by Apache Spark
  • IBM Knowledge Catalog
None
Data Refinery None
None
None
Data Replication None
None
The service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
DataStage® None
The following services are not required but provide additional functionality:
  • Watson Pipelines enables you to convert DataStage sequence jobs to pipelines.
The service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
Db2® None
The following services are not required but provide additional functionality:
  • Db2 Data Management Console provides:
    • A graphical user interface for SQL execution
    • A runtime monitoring interface
This service automatically installs the following dependencies if they are not already installed:
  • Db2U (db2u)
Db2 Big SQL To use Db2 Big SQL, you must have remote data storage, such as:
  • A Hadoop cluster (Cloudera Data Platform Version 7.1.7)
  • Object storage
    Db2 Big SQL supports:
    • IBM Cloud Object Storage
    • Amazon Web Services object storage
    • IBM Storage Scale object storage
    • Red Hat OpenShift Data Foundation
The following services are not required but provide additional functionality:
  • Db2 Data Management Console provides:
    • A graphical user interface for SQL execution
    • A runtime monitoring interface
This service automatically installs the following dependencies if they are not already installed:
  • Db2U (db2u)
Db2 Data Gate To use this service, you must have:
  • IBM z/OS® V2.4 (5650-ZOS) or later.
  • Db2 for z/OS, either:
    • V12 (5650-DB2 or 5770-AF3) with APAR fixes installed and running at Function Level 505 or higher. You find the list of required APAR fixes here: Software dependencies.
    • V13 (5698-DB2 or 5698-DBV)
  • Distributed data facility (DDF) with a secure port, configured for network encryption through AT-TLS. For details, see Configuring network access between Db2 Data Gate and IBM Z®.
To use this service, you must have at least one instance of a supported, integrated database:
  • Db2
  • Db2 Warehouse
The following services are not required but provide additional functionality:
  • IBM Knowledge Catalog enables you to automatically publish metadata about Db2 Data Gate tables to catalogs.
None
Db2 Data Management Console None
None
This service automatically installs the following dependencies if they are not already installed:
  • Redis (ibm_redis_cp)
Db2 Warehouse None
The following services are not required but provide additional functionality:
  • Db2 Data Management Console provides:
    • A graphical user interface for SQL execution
    • A runtime monitoring interface
This service automatically installs the following dependencies if they are not already installed:
  • Db2U (db2u)
Decision Optimization None
To install this service, you must have the following services already installed:
  • Watson Studio
  • Watson Machine Learning
None
EDB Postgres None
None
This service automatically installs the following dependencies if they are not already installed:
  • Cloud Native PostgreSQL (postgresql)
Execution Engine for Apache Hadoop To use this service, you must have an Execution Engine for Apache Hadoop RPM installation on a Hadoop cluster. 
To install this service, you must have the following services already installed:
  • Watson Studio
None
IBM Knowledge Catalog None
The service automatically installs the following services if they are not already installed:
  • Analytics Engine powered by Apache Spark
  • Data Refinery
If you choose to install the data quality feature, the service automatically installs the following services if they are not already installed:
  • DataStage Enterprise
The following services are not required but provide additional functionality:
  • Data Privacy
  • MANTA Automated Data Lineage
The service automatically installs the following dependencies if they are not already installed:
  • Common core services
  • Db2U (db2u)
  • Db2 as a service (db2aaservice)
If you choose to install the semantic search and data lineage feature, the service automatically installs the following dependencies if they are not already installed:
  • FoundationDB (opencontent_fdb)
IBM Match 360 with Watson None
The following service is not required but provides additional functionality:
  • IBM Knowledge Catalog enables key IBM Match 360 capabilities such as profiling, automapping, data quality workflows, and data governance.
This service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
  • OpenSearch (opencontent_elasticsearch)
  • FoundationDB (opencontent_fdb)
  • RabbitMQ (opencontent_rabbitmq)
  • Redis (opencontent_redis)
Informix® None
None
This service automatically installs the following dependencies if they are not already installed:
  • informix
MANTA Automated Data Lineage None
To install this service, you must have the following services already installed:
  • IBM Knowledge Catalog
None
MongoDB None
None
You must install the following dependencies when you install the service:
  • mongodb
OpenPages If you prefer to connect to an external database rather than having OpenPages automatically provision a Db2 database for you, you must have IBM Db2 on Linux.
The following services are not required but provide additional functionality:
  • AI Factsheets enables you to review machine learning models and related activities as part of enterprise risk and compliance monitoring.
  • Cognos Analytics
  • watsonx Assistant
  • Watson Discovery
  • IBM Knowledge Catalog
  • Watson OpenScale
This service automatically installs the following dependencies if they are not already installed:
  • Db2U (db2u)
  • Db2 as a service (db2aaservice)
  • RabbitMQ (opencontent_rabbitmq)
Planning Analytics If you want to use Microsoft Excel, you must have the Planning Analytics for Microsoft Excel plug-in.
None
None
Product Master If you want to connect to a database outside of Cloud Pak for Data, it must be an IBM Db2 or Oracle database.
If you want to connect to an integrated database, you must have the following service already installed:
  • Db2
None
RStudio® Server Runtimes None
To install this service, you must have the following service already installed:
  • Watson Studio
None
SPSS® Modeler None
To install this service, you must have the following services already installed:
  • Watson Studio
This service automatically installs the following dependencies if they are not already installed:
  • Canvas (canvasbase)
Synthetic Data Generator None
None
 This service automatically installs the following dependencies if they are not already installed:
  • Canvas (canvasbase)
  • Common core services (ccs)
Voice Gateway None
To install this service, you must have the following service already installed:
  • watsonx Assistant
  • Watson Speech to Text
  • Watson Text to Speech
None
Watson Discovery To install this service, you must have Multicloud Object Gateway. For more information, see Installing and setting up Multicloud Object Gateway
None
This service automatically installs the following dependencies if they are not already installed:
  • Cloud Native PostgreSQL (postgresql)
  • etcd (opencontent_etcd)
  • Elasticsearch (opencontent_elasticsearch)
  • MinIO (opencontent_minio)
  • RabbitMQ (opencontent_rabbitmq)
  • Watson Gateway (watson_gateway)
  • Watson model trainer (model_train)
Watson Machine Learning None
The following services are not required but provide additional functionality:
  • Watson Machine Learning Accelerator provides the Experiment Builder user interface.
The service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
Watson Machine Learning Accelerator To install this service, you must install the NVIDIA GPU Operator:
x86-64
  • On OpenShift 4.12, use v23.9.1, v23.9.0, v23.6.1, v23.3.2, v22.9.2
  • On OpenShift 4.14, use v23.9.1, v23.9.0
To install this service, you must have the following service already installed:
  • Scheduling service
The following services are not required but provide additional functionality:
  • Watson Machine Learning enables you to use Deep Learning Experiments
None
Watson OpenScale If you prefer to connect to an external database, you must have Db2 Enterprise Server Edition 11.5 or later.
If you want to connect to an integrated database, you must have at least one instance of a supported, integrated database:
  • Db2
  • Db2 Warehouse
  • EDB Postgres

The following services are not required but provide additional functionality. To use the functionality, you must install the services before you install Watson OpenScale:

  • Watson Studio enables you to:
    • Create AutoAI models and Jupyter Notebooks
    • Set up a demo environment where you can quickly tour the Watson OpenScale capabilities
  • Watson Machine Learning enables you to:
    • Create a deployed model that Watson OpenScale can check for bias and drift
    • Automatically log payloads
None
Watson Pipelines None
The following services are not required but provide additional nodes when they are installed:
  • DataStage
  • Watson Machine Learning
The service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
Watson Query None
The service automatically installs the following services if they are not already installed:
  • Db2 Data Management Console
The service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
  • Db2U (db2u)
Watson Speech services To install this service, you must have Multicloud Object Gateway. For more information, see Installing and setting up Multicloud Object Gateway
None
This service automatically installs the following dependencies if they are not already installed:
  • Cloud Native PostgreSQL (postgresql)
  • MinIO (opencontent_minio)
  • RabbitMQ (opencontent_rabbitmq)
  • Watson Gateway (watson_gateway)
Watson Studio None
The service automatically installs the following services if they are not already installed:
  • Data Refinery
  • Watson Studio Runtimes
This service automatically installs the following dependencies if they are not already installed:
  • Common core services (ccs)
Watson Studio Runtimes To install the following environments, you must have the NVIDIA GPU Operator installed:
  • Runtime 22.2 on Python 3.10 for GPU
  • Runtime 23.1 on Python 3.10 for GPU
The following versions of the NVIDIA GPU Operator are supported:
x86-64
  • On OpenShift 4.12, use v23.9.1, v23.9.0, v23.6.1, v23.3.2, v22.9.2
  • On OpenShift 4.14, use v23.9.1, v23.9.0
To install this service, you must have the following service already installed:
  • Watson Studio
None
watsonx Assistant To install this service, you must have:
The following services are not required but provide additional functionality:
  • Watson Discovery enables you to add a search skill to your assistant.
This service automatically installs the following dependencies if they are not already installed:
  • Cloud Native PostgreSQL (postgresql)
  • etcd (opencontent_etcd)
  • Elasticsearch (opencontent_elasticsearch)
  • MinIO (opencontent_minio)
  • Redis (ibm_redis_cp)
  • Watson data governor (data_governor)
  • Watson Gateway (watson_gateway)
watsonx.ai
To install this service, you must install the following operators:
  • NVIDIA GPU Operator v23.61
    Note: The Multi-Instance GPU (MIG) feature of the NVIDIA GPU must be disabled.
  • Node Feature Discovery Operator
    Install the stable version of the operator. For more information, see Node Feature Discovery Operator in the Red Hat OpenShift Container Platform documentation:

You can optionally install Red Hat OpenShift AI Version 2.6. For more information see the Product Documentation for Red Hat OpenShift AI Self-Managed.
The following services are not required but provide additional functionality:
  • watsonx.governance enables you to govern your generative AI assets.
 None
watsonx.data None
  • 4.8.1 4.8.2 4.8.3 4.8.4 The service does not have any dependencies on other services.
  • 4.8.5 or later The service automatically installs the following services if they are not already installed:
    • Analytics Engine powered by Apache Spark
 This service automatically installs the following dependencies if they are not already installed:
  • Cloud Native PostgreSQL (postgresql)
watsonx.governance None
If you want to connect to an integrated database for watsonx.governance's Model Management offering, you must have at least one instance of a supported, integrated database:
  • Db2
  • Db2 Warehouse
  • EDB Postgres
The following services are not required but provide additional functionality:
  • Cognos Analytics enables you to generate reports and create dashboards.
  • watsonx.ai enables you to build and deploy generative AI assets.
 This service automatically installs the following dependencies if they are not already installed:
  • Db2U (db2u)
  • Db2 as a service (db2aaservice)
  • Elasticsearch (opencontent_elasticsearch)
  • RabbitMQ (opencontent_rabbitmq)
watsonx Orchestrate®
To install this service, you must install IBM App Connect in containers on the cluster. For more information, see Installing IBM App Connect in containers.

You can optionally install IBM Robotic Process Automation if you want to run Robotic Process Automation bots as skills in watsonx Orchestrate. For more information, see Installing IBM Robotic Process Automation.
To install this service, you must have the following service already installed:
  • watsonx Assistant
 This service automatically installs the following dependencies if they are not already installed:
  • Cloud Native PostgreSQL (postgresql)
  • MongoDB (both mongodb and mongodb_cp4d)
  • RabbitMQ (opencontent_rabbitmq)
  • Redis (ibm_redis_cp)

Supported web browsers

You can use the following web browsers to access the Cloud Pak for Data web client.

  • Mozilla Firefox (recommended)
  • Google Chrome
  • Microsoft Edge

It is recommended that you use the latest available version or the latest version approved by your enterprise.