Preparing to install Watson Knowledge Catalog

Plan and prepare to install Watson™ Knowledge Catalog.

Before you begin

Complete these tasks before you install Watson Knowledge Catalog.

Determining which configuration to install

You need to decide which configuration of Watson Knowledge Catalog you want to install.

Watson Knowledge Catalog has two main configurations: the base configuration and the core configuration.

Important: Starting in version 4.6.4, the core version is the default configuration that is installed. For earlier versions, the base configuration is installed by default.

The base configuration includes legacy features that are based on Information Server components. Some legacy features require permissions that are not included in the predefined roles. Many of the legacy features are replaced with new equivalent features that are available in both the base and core configurations. Specify the core configuration if you do not need the legacy features and want a smaller installation size.

For more information, see Specifying installation options.

Determining the optional features to enable

You need to decide which optional features, if any, you want to enable for Watson Knowledge Catalog. These features are enabled through the Watson Knowledge Catalog custom resource (CR) during install or upgrade. For more information on how to enable these during install or upgrade, see Installing Watson Knowledge Catalog or Upgrading Watson Knowledge Catalog.

These optional features can also be enabled after the initial install or upgrade. If you decide to enable them after, see Enabling optional features after installation or upgrade for Watson Knowledge Catalog.

Use the following table to decide which optional features you would like to enable and find out which optional component you would need to enable that feature.
Table 2. Optional features and the component needed to enable
Feature Component needed Entry in the custom resource
Lineage Knowledge graph
enableKnowledgeGraph: True
Semantic search Knowledge graph
enableKnowledgeGraph: True
Data quality features in analytics projects Data quality
enableDataQuality: True
Advanced metadata import MANTA Automated Data Lineage
enableMANTA: True

Configuring Db2 on a specific node

You can specify Db2® to run on specific nodes of your cluster. You might want to make this specification if, for example, your environment does not allow custom security context constraints (SCCs) to be deployed on certain nodes.

Important: You must complete this task when you create the Watson Knowledge Catalog custom resource (CR) at the time of installation. You cannot set these configurations after Watson Knowledge Catalog is installed.
  1. Create a custom machine configuration pool (MCP) by following the steps at Custom Machine Config Pool in OpenShift 4.
  2. Assign the KubeletConfig file to the node that you want by using the custom MCP.
  3. Label the node in the following format:
    oc label node <node> icp4data=<label>
    See the following example:
    oc label node worker1.newco-cpd.cp.ibm.com icp4data=wkcdb2u

    For more information, see Setting up dedicated nodes for your Db2 deployment.

  4. Skip this step until you have created the Watson Knowledge Catalog custom resource (CR) during the initial install process. Then, add the following override to the Watson Knowledge Catalog CR:
    spec:
     wkc_db2u_node_label: wkcdb2u
     iis_db2u_node_label: iisdb2u