IBM Data and AI

Live Demos as a Service

This run book outlines the steps to setup and configure Cloud Pak for Data as a Service on IBM Cloud to support the IBM Data and AI Live Demos demonstrations and labs. The documentation applies to the production environments with a few exceptions, which are noted in the instructions.

The tasks are organized by persona but you will perform all the tasks as the Data Steward administrator to setup, configure, and create all Cloud Pak for Data cloud services and objects (Platform connections, Virtualized data Sources and data, Catalogs, Business Glossary, Policies and rules etc.) This ensures that all services and objects are owned and administered by the Data Steward user who holds administration privileges.

Cloud Account Management

1. Login to IBM Cloud

The production cloud acount is named Techzone Outcomes. Follow the instructions in the following sections to login to the appropriate cloud account using the supplied credentials. The Data Steward account is the administrator account for all environments.

Production Data Steward Account

Login in to the Techzone Outcomes cloud account as the Data Steward administrator using the following credentials:

Login id = Your data steward Login Id

Password = Your data steward password

API Key = Your data steward Id API Key

Production Business User Account

The Business User account is available in production to simulate what end users experience and to validate the demos and labs.

Login Id = Your business user Login Id

Password = Your business user password

Cloud Account API Keys

The API keys serve as the credentials for the Watson Query connection that is created in the Governance catalog that houses all governed virtualized data. It is also the key used by the outcomes reservation system to create and invite users, assign them to the Business User data access group and to assign Viewer service privileges for Watson Query, Watson Knowledge Catalog, Watson Studio and Watson OpenScale.

TechZone Outcomes

Data Steward API Key = Your data steward API Key

Business User API Key = Your business user API Key

2. Create Cloud Resource Groups

Cloud Pak for Data Outcomes

Create the following Cloud resource groups. This will be the resource group you select when creating all of the Cloud Pak for Data services in the upcoming sections, except for the Outcomes Storage service, defined in the next step.

1. Select the Manage menu from the IBM Cloud tool bar.

2. Select the Account menu item.

3. Click Resource Groups from the left side menu.

4. Click Create.

5. Enter Name = Cloud Pak for Data Outcomes.

6. Click Create.

Outcomes Project Management

Create the following Cloud resource group. It will be the resource group you select when creating the Outcomes Storage cloud object storage service that will house all of our catalogs; Business, Governance and Platform assets catalog, and our projects that we use to build out the environment, the Outcomes Runbook and Outcomes Catalog projects.

1. Select the Manage menu from the IBM Cloud tool bar.

2. Select the Account menu item.

3. Click Resource Groups from the left side menu.

4. Click Create.

5. Enter Name = Outcomes Project Management.

6. Click Create.

3. Create Cloud Services

Create the following Cloud Pak for Data services from the IBM Cloud catalog, as described below, by typing the name of the service into the service catalog search area, selecting the service and then creating the service with the designated plan specified:

Outcomes Cloud Object Storage

This cloud object storage service is used by the Data and AI team to create objects that need to be isolated from end users like our environment configuration projects, and all catalogs. This COS service is not visible or accessible to end users.

• Name = Outcomes Storage

• Plan = Standard

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

This cloud object storage service is used by all end users to create their projects and is the only cloud object storage that is visible for them to select when creating projects.

Demo Cloud Object Storage

• Name = Demo Storage

• Plan = Standard

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Data Replication

• Name = Data Replication

• Plan = Lite

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

DataStage

• Name = DataStage

• Plan = Standard

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Databases for DataStax

• Name = Databases for DataStax

• RAM = 24 GB

• Disk = 360 GB

• Cores = 6

• Members = 3

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Databases for EDB

• Name = Databases for EDB

• RAM = 10 GB

• Disk = 100 GB

• Cores = 3

• Members = 3

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Databases for MongoDB

• Name = Databases for MongoDB

• Plan = Standard

Make sure you choose standard and not enterprise. The Enterprise edition uses SCRAM-> SHA-256 authentication and standard uses SCRAM SHA-1 and the CP4D MongoDB > driver only supports SCRAM-SHA-1.

• RAM = 24 GB

• Disk = 120 GB

• Cores = 6

• Members = 3

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Db2 Warehouse

• Name = Db2 Warehouse

• Plan = Flex One

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

IBM Cognos Dashboard Embedded

• Name = IBM Cognos Dashboard

• Plan = Pay as you go

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

IBM Match 360 for Watson

• Name = IBM Match 360

• Plan = Lite

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Machine Learning

• Name = Machine Learning

• Plan = v2 Professional

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Speech to Text

• Name = Speech to Text

• Plan = Plus

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Text to Speech

• Name = Text to Speech

• Plan = Standard

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Watson Assistant

• Name = Watson Assistant

• Plan = Plus

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Watson Discovery

• Name = Watson Discovery

• Plan = Enterprise

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Watson Knowledge Catalog

• Name = Watson Knowledge Catalog

• Plan = Enterprise

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Watson OpenScale

• Name = Watson OpenScale

• Plan = Standard v2

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Watson Studio

• Name = Watson Studio

• Plan = Enterprise v2

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

Watson Query

• Name = Watson Query

• Plan = Enterprise

• Resource Group = Cloud Pak for Data Outcomes

• Tags = cpdaas and outcomes

4. Change Cloud Database Passwords

The cloud databases that were just created; Databases for DataStax, Databases for EDB and Databases for MongoDB, need to have their administrator passwords changed to administer them properly. Follow the steps below to set the admin password for these services:

1. Go to the IBM Cloud account left side menu.

2. Select Resource List.

3. Open the Services and Software category.

Databases for DataStax Outcomes

1. Click on the Databases for DataStax Outcomes service.

2. Select the Settings tab.

3. Scroll down to the Change Database Admin Password section.

4. Copy and paste this password - Your admin password into the New Password field.

5. Click Change Password.

6. Click the back button on your browser to go back to the resource list.

Databases for EDB Outcomes

1. Click on the Databases for EDB Outcomes service.

2. Select the Settings tab.

3. Scroll down to the Change Database Admin Password section

4. Copy and paste this password - Your admin password into the New Password field.

5. Click Change Password.

6. Click the back button on your browser to go back to the resource list.

Databases for MongoDB Outcomes

1. Click on the Databases for DataStax Outcomes service.

2. Select the Settings tab.

3. Scroll down to the Change Database Admin Password section

4. Copy and paste this password - Your admin password into the New Password field.

5. Click Change Password.

6. Click the back button on your browser to go back to the resource list.

Setup Cloud Account Security

Perform the following tasks as the Data Steward cloud account administrator to create the data access groups, and the Identify Access Management (IAM) policies that are needed for end users to perform the Cloud Pak for Data - Outcomes demonstrations:

1. Go to the IBM Cloud toolbar.

2. Select Manage > Access (IAM).

1. Create Business User Data Access Group

1. Select the Access groups menu item from the left side menu.

2. Click Create.

3. Enter a Name = Business User

4. Enter a Description = This group includes all users who have requested access to the Cloud Pak for Data Outcomes demo environment. Users in this group are given read only access to the Cloud Pak for Data services and capabilities that are used in the demonstrations. (including the period at the end).

5. Click Create.

2. Create Administrators Data Access Group

1. Select the Access groups menu item from the left side menu.

2. Click Create.

3. Enter a Name = Administrators

4. Enter a Description = This group includes all users who have admin access to the Cloud Pak for Data Outcomes demo environment. Users in this group are given admin access to the Cloud Pak for Data services and capabilities that are used in the demonstrations. (including the period at the end).

5. Click Create.

3. Create Service to Service Authorization

In order for Watson Knowledge Catalog and Watson Query to communicate and share a catalog for WQ to publish to, and for WQ to enforce WKC data protection rules, a service to service authorization has to be created in IAM.

Follow the steps below to establish the authorization:

1. Select the Authorizations menu item from the left side menu.

2. Click Create.

3. Select the Watson Knowledge Catalog service from the Source service dropdown list.

4. Select the Resources based on selected attributed radio button.

5. Select the Source resource group checkbox.

6. Select the Cloud Pak for Data resource group from the dropdown list.

7. Select Watson Query from the Target service dropdown list.

8. Select the Resources based on selected attributed radio button.

9. Select the Source resource group checkbox.

10. Select the Cloud Pak for Data resource group from the dropdown list.

11. Click the checkbox in the Service access section next to DataAccess (For Service to Service Authorization Only).

12. Click Authorize.

4. Create Watson Knowledge Catalog Business User Role

1. Select the Roles menu item from the left side menu.

2. Click Create.

3. Enter a Name = Watson Knowledge Catalog Business User

4. Enter a ID = WKCUser

5. Enter a Description = Permissions for business users to use the Watson Knowledge Catalog service. (including the period at the end).

6. Select IBM Cloud Pak for Data from the service dropdown list.

7. Select All roles from the view actions for dropdown list.

8. Click the Add link next to the following actions:

   • cp4d.catalog.access

   • cp4d.governance-artifacts.access

9. Click Apply.

5. Create Watson Knowledge Catalog Administrator Role

1. Select the Roles menu item from the left side menu.

2. Click Create.

3. Enter a Name = Watson Knowledge Catalog Administrator

4. Enter a ID = WKCAdmin

5. Enter a Description = Permissions for administrators of the Watson Knowledge Catalog service. (including the period at the end).

6. Select IBM Cloud Pak for Data from the service dropdown list.

7. Select All roles from the view actions for dropdown list.

8. Click the Add link next to the following actions:

   • cp4d.catalog.access

   • cp4d.catalog.manage

   • cp4d.data-protection-rules.manage

   • cp4d.governance-artifacts.access

   • cp4d.governance-categories.manage

   • cp4d.governance-workflows.manage

   • cp4d.wkc.reporting.manage

9. Click Apply.

6. Assign IAM Policies to Business User Group

Business users, users that have requested access to the demo environment using the Cloud Pak for Data - Outcomes reservation system, are given view only access to platform and service capabilities.

Assign IAM Services Roles

1. Select the Access groups menu item from the left side menu.

2. Click the Business User group.

3. Click the Access policies tab.

4. Click the Assign access button.

5. Click the IAM services tile.

6. Select All Identity and Access Enabled services from the Which service do you want to assign access to? dropdown list.

7. Click the Resources based on selected attributes radio button.

8. Click the Resource group checkbox.

9. Select the Cloud Pak for Data resource group from the dropdown list.

10. Check the Viewer checkbox for Platform access.

11. Check the Reader checkbox for Service access.

12. Check the Viewer checkbox for Resource group access.

13. Click Add.

14. Click Assign.

Assign Watson Knowledge Catalog User Role

1. Click the Assign access button.

2. Click the IAM services tile.

3. Select IBM Cloud Pak for Data from the Which service do you want to assign access to? dropdown list.

4. Scroll down to the bottom of the page to the Custom access section.

5. Click the Watson Knowledge Catalog User checkbox.

6. Click Add.

7. Click Assign.

Assign Machine Learning Role

1. Click the Assign access button.

2. Click the IAM services tile.

3. Select Machine Learning from the Which service do you want to assign access to? dropdown list.

4. Click the Resources based on selected attributes radio button.

5. Click the Service instance checkbox.

6. Select string equals from the first dropdown list box.

7. Select the Machine Learning Outcomes (b7b...) instance from the service dropdown list.

   • There should only be one Machine Learning Outcomes instance in the list.

8. Check the Editor checkbox for Platform access.

9. Check the Writer checkbox for Service access.

10. Click Add.

11. Click Assign.

7. Assign IAM Policies to Administrator Group

Administrators, users that have assigned administrative privileges to the Cloud Pak for Data - Outcomes environment, are given full access to platform and service capabilities.

Assign IAM Services Roles

1. Select the Access groups menu item from the left side menu.

2. Click the Administrator group.

3. Click the Access policies tab.

4. Click the Assign access button.

5. Click the IAM services tile.

6. Select All Identity and Access Enabled services from the Which service do you want to assign access to? dropdown list.

7. Click the All resources radio button.

8. Check the Administrator checkbox for Platform access.

9. Check the Manager checkbox for Service access.

10. Check the Administrator checkbox for Resource group access.

11. Click Add.

12. Click Assign.

Assign Account Management Roles

1. Click the Assign access button.

2. Click the Account management tile.

3. Select All Account Management services from the Which service do you want to assign access to? dropdown list.

4. Check the Administrator checkbox for Platform access.

5. Click Add.

6. Click Assign.

Cloud Pak for Data Tasks

1. Login to Cloud Pak for Data

The following tasks are done from the Cloud Pak for Data as a service application using the Techzone Outcomes cloud account as the Data Steward administrator.

Make sure you are in the Techzone Outcomes cloud account by ensuring it is selected in the top right corner of the UI.

Go to the Cloud Pak for Data as a Service home page and login with the Data Steward credentials provided in the first section of this document.

2. Enable Storage Delegation

1. Go to the Cloud Pak for Data navigation menu.

2. Select Administration > Storage delegation.

Perform the following steps for the Demo Storage instance:

1. Click the Projects button and set it to the green, on position.

2. Click the Catalogs button and set it to the green, on position.

Note: - DO NOT turn these flags on for the Outcomes Storage instance. They should remain turned off.

3. Create Catalogs

This section creates three catalogs; The Business catalog, Platform assets catalog, and the Governance catalog.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Catalogs > View all catalogs.

Create Business Catalog

This catalog contains all the governed and trusted data assets that are used by the business for analytics and AI. All users that are assigned to the Business User group are given Viewer access to this catalog.

1. Click Create Catalog.

2. Enter Name = Business

3. Enter Description = This catalog stores governed assets used by the business for analytical and AI projects.. (including the period at the end).

4. Check the box to Enforce data policies.

5. Click OK when prompted if you are sure you want to enforce policies.

6. Select the Update original assets duplicate asset handling option.

7. Select the Cloud Object Storage instance from the Object storage instance dropdown.

8. Click Create.

9. Select the Access control tab.

10. Scroll down and select Add access groups.

11. Select Viewer for the access level.

12. Start typing bus in the search area.

13. Click the Business Users access group.

14. Click Add.

Create Governance Catalog

This catalog is only used by Data Virtualization and has no user access. Only the Data Steward user, assigned as the administrator, will have access to this catalog.

1. Click Create Catalog.

2. Enter Name = Governance

3. Enter Description = This catalog stores governed data assets that are virtualized through Data Virtualization. (including the period at the end)

4. Check the box to Enforce data policies.

5. Click OK when prompted if you are sure you want to enforce policies.

6. Select the Update original assets duplicate asset handling option.

7. Select the Cloud Object Storage instance from the Object storage instance dropdown.

8. Click Create.

Create Platform assets catalog

You must create the Platform assets catalog before you can create platform connections. This is only done once to allow for platform connections to be shared across the platform.

Users will only have read only access to the platform connections that are published to this catalog.

Note: This catalog is not used or serve any purpose for Cloud Pak for Data - Outcomes demonstrations and labs, but is required for users to access and view Platform connections from the main menu.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Data > Platform connections.

3. Click Create Catalog.

4. Select the Cloud Object Storage instance from the Object storage instance dropdown.

5. Click Create.

6. Select the Access control tab.

7. Scroll down and select Add access groups.

8. Select Viewer for the access level.

9. Start typing bus in the search area.

10. Click the Business Users access group.

11. Click Add.

Create Governance Artifacts

This section creates all the governance artifacts needed to establish a data governance foundation. It builds out a fully published business glossary and defines and enforces data governance policies and rules. It uses a set of CSV files to import and create the artifacts.

The files are located in this Cloud Pak for Data - Outcomes GitHub repository:

Download and unzip the cpd-outcomes-governance.zip file to a location on your desktop and remember where you put it. You will be instructed to select specific files that are contained in this file during the creation of the governance artifacts in the upcoming steps.

1. Create Categories

Categories have to be created before any other governance artifacts that require an assignment to a category are created. The only governance artifact that does not require an association to a category are data protection rules. Categories are like folders and are a high level umbrella and container for other governance artifacts. Categories also control all governance artifact access and security levels.

The import should succeed with 14 new categories. If the categories do not appear, go to the navigation main menu and select Governance > Categories.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Governance > Categories.

3. Select Add category > Import from file.

4. Select Add file.

5. Find and select the governance-categories.csv file from your download location.

6. Click Open.

7. Click Next.

8. Select Replace all values.

9. Select Import.

10. Select Close.

11. Select the Refresh button on your browser.

12. Select the category explorer icon in the top left corner to get a tree view.

The import should succeed with 14 new categories. If the categories do not appear, go to the navigation main menu and select Governance > Categories.

Select the category explorer icon in the top left corner. You should see 9 primary categories and 5 sub-categories:

• Banking

  • Mortgage

• Customer

• Data Privacy

  • Payment Card Industry

  • Personal Information

  • Personally Identifiable Information

  • Sensitive Personal Information

• Employee

• Insurance

• Location

• Person

• Transportation

• Universal

2. Update Classifications

1. Go to the Cloud Pak for Data navigation menu

2. Select Governance > Classifications.

3. Select Add classification > Import from file.

4. Click Add file.

5. Select the governance-classifications.csv file from your download location.

6. Click Open.

7. Click Next.

8. Select Replace all values.

9. Click Import.

10. Click Go to Task.

11. Click Publish.

12. Go to the Cloud Pak for Data navigation menu.

13. Select Governance > Classifications**.

You should see 4 new published classifications that are assigned to the Data Privacy category along side the 4 already published classifications of the same name that are assigned to the [uncategorized] category:

• Confidential

• Personal Information

• Personally Identifiable Information

• Sensitive Personal Information

The duplicate classifications assigned to the [uncategorized] category are no longer needed so we will delete them.

1. Click the Edit button.

   Looks like a pencil icon next to the Sort criteria to the far right of the screen

2. Select the check box next to all classifications that are assigned to the [uncategorized] category.

3. Scroll up to the top of the screen.

4. Select Mark for deletion.

You will see a dialog box indicating it was submitted for deletion. Wait a few seconds for the second dialog box to appear that has a link to the task you need to process.

6. Select the task link in the second dialog box.

   • This will take you directly to the Mark for deletion task.

   • If you miss the chance to click on the link from the dialog box:

     • Go to the main navigation menu and select Governance > Task inbox.

     • Select the checkbox next to the classification deletion task.

7. Click Delete.

8. Go to the Cloud Pak for Data navigation menu.

9. Select Governance > Classifications.

You should now see 4 published Classifications assigned to the Data Privacy category.

3. Create Data Classes

1. Go to the Cloud Pak for Data navigation menu.

2. Select Governance > Data classes.

3. Click New data class.

4. Click Import from file.

5. Click Add file.

6. Select the governance-data-classes.csv file from your download location.

7. Click Open.

8. Click Next.

9. Select Replace all values.

10. Click Import.

11. Click Go to Task.

You should see 15 new data classes to publish.

12. Click Publish.

13. Go to the Cloud Pak for Data navigation menu.

14. Select Governance > Data classes

15. Click the Sort by dropdown list.

16. Select Last modified .

There should be 15 newly published data classes that are assigned to the [uncategorized] category that have a modification date of today.

4. Create Business Terms

Business terms have references to data classes and classifications and are referenced in governance rules, so they need to be imported after data classes and classifications but before importing or creating governance rules.

Business terms can also be dependent on other business terms which establishes a parent child relationship between them. Therefore, we will import the parent business terms before importing all the other business terms.

There should be NO business terms in the business glossary at this point.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Governance > Business terms.

Import Parent Business Terms

1. Select Add business term > Import from file.

2. Click Add file.

3. Select the governance-business-terms-parent.csv file from the download location.

4. Click Open.

5. Click Next.

6. Select Replace all values.

7. Click Import.

8. Click Go to Task.

9. Click Publish.

10. Go to the Cloud Pak for Data navigation menu.

11. Select Governance > Business terms.

All business terms should be appearing as Published. Verify that there are 6 published business terms before proceeding to the next step because some of the business terms in the next import are dependent on them being published first.

Import Child Business Terms

1. Select Add business term > Import from file.

2. Click Add file.

3. Select the governance-business-terms.csv file from the download location.

4. Click Open.

5. Click Next.

6. Select Replace all values.

7. Click Import.

8. Click Go to Task.

There should be 328 business term drafts waiting to be published...

9. Click Publish.

10. Go to the Cloud Pak for Data navigation menu.

11. Select Governance > Business terms.

All business terms should be appearing as Published. Scroll to the bottom of the business term list and verify that there are now 328 published business terms.

5. Create Reference Data

The CSV files for reference data are included as part of the cpd-outcomes-governance-artifacts.zip file you downloaded and unzipped earlier.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Governance > Reference.

Create the following reference data sets:

Department Lookup

1. Select Add reference data set > New reference data set.

2. Select the DEPARTMENT_LOOKUP.csv as the file to upload.

3. Enter Department Lookup (case sensitive) as the reference data name.

4. Select Text as the reference data type.

5. Click the Change category button.

6. Select the Employee category.

7. Click Add.

8. Copy and pastes this description: Valid codes and values for all company departments. including the period at the end).

9. Click Next.

10. Make sure the First row as column header is set to On.

11. Click Select column dropdown for the DEPARTMENT_CODE target column.

12. Select Code.

13. Click Select column dropdown for the DEPARTMENT_EN target column.

14. Select Value.

15. Click Next.

16. Click Create.

From the About this reference data panel on the right:

1. Click on the plus sign + next to Tags.

2. Click the dropdown arrow in the search tags area.

3. Select the Employee tag.

4. Click Done.

5. Click Publish.

6. Click Publish.

You will see an informational icon and message of Draft preview This artifact has a new published version.

1. Click Reload artifact.

2. Click the Reference data bread crumb in the top left corner.

   This will take you back to the Reference data home page.

Gender Lookup

1. Select Add reference data set > New reference data set.

2. Select the GENDER_LOOKUP.csv as the file to upload.

3. Enter Gender Lookup (case sensitive) as the reference data name.

4. Select Text as the reference data type.

5. Click the Change category button.

6. Select the Person category.

7. Click Add.

8. Copy and pastes this description: Valid codes and values for an individual's gender. (including the period at the end).

9. Click Next.

10. Make sure the First row as column header is set to On.

11. Click Select column dropdown for the GENDER_CODE target column.

12. Select Code.

13. Click Select column dropdown for the GENDER_EN target column.

14. Select Value.

15. Select Next.

16. Click Create.

From the About this reference data panel on the right:

1. Click on the plus sign + next to Tags.

2. Click the dropdown arrow in the search tags area.

3. Select the Customer tag.

4. Select the Employee tag.

5. Type Person in the tag search area and Select the Person tag.

6. Click Done.

7. Click Publish.

8. Click Publish.

You will see an informational icon and message of Draft preview This artifact has a new published version.

1. Click Reload artifact.

2. Click the Reference data bread crumb in the top left corner.

   • This will take you back to the Reference data home page.

Position Lookup

1. Select Add reference data set > New reference data set.

2. Select the POSITION_LOOKUP.csv as the file to upload.

3. Enter Position Lookup (case sensitive) as the reference data name.

4. Select Text as the reference data type.

5. Click the Change category button.

6. Select the Employee category.

7. Click Add.

8. Copy and pastes this description: Valid codes and values for positions across the company. (including the period at the end).

9. Click Next.

10. Make sure the First row as column header is set to On.

11. Click Select column dropdown for the POSITION_CODE target column.

12. Select Code.

13. Click Select column dropdown for the POSITION_EN target column.

14. Select Value.

15. Click Next.

16. Click Create.

From the About this reference data panel on the right:

1. Click on the plus sign + next to Tags.

2. Click the dropdown arrow in the search tags area.

3. Select the Employee tag.

4. Click Done.

5. Click Publish.

6. Click Publish.

You will see an informational icon and message of Draft preview This artifact has a new published version.

1. Click Reload artifact.

2. Click the Reference data bread crumb in the top left corner.

   • This will take you back to the Reference data home page.

Termination Lookup

1. Select Add reference data set > New reference data set.

2. Select the TERMINATION_LOOKUP.csv as the file to upload.

3. Enter Termination Lookup (case sensitive) as the reference data name.

4. Select Text as the reference data type.

5. Click the Change category button.

6. Select the Employee category.

7. Click Add.

8. Copy and pastes this description: Valid codes and values for employee terminations. (including the period at the end).

9. Click Next.

10. Make sure the First row as column header is set to On.

11. Click Select column dropdown for the TERMINATION_CODE target column.

12. Select Code.

13. Click Select column dropdown for the TERMINATION_REASON_EN target column.

14. Select Value.

15. Click Next.

16. Click Create.

From the About this reference data panel on the right:

1. Click on the plus sign + next to Tags.

2. Click the dropdown arrow in the search tags area.

3. Select the Employee tag.

4. Click Done.

5. Click Publish.

6. Click Publish.

You will see an informational icon and message of Draft preview This artifact has a new published version.

1. Click Reload artifact.

2. Click the Reference data bread crumb in the top left corner.

   • This will take you back to the Reference data home page.

You should see 4 Reference data assets:

• Department Lookup

• Gender Lookup

• Position Lookup

• Termination Lookup

6. Create Governance Rules

1. Go to the Cloud Pak for Data navigation menu.

2. Select Governance > Rules.

3. Select Add rule > Import from file.

4. Click Add file.

5. Select the governance-rules.csv file from the download location.

6. Click Open.

7. Click Next.

8. Select Replace all values.

9. Click Import.

10. Click Go to Task.

11. Click Publish.

12. Go to the Cloud Pak for Data navigation menu.

13. Select Governance > Rules.

You should see 4 published governance rules with names that start with All...

7. Create Data Protection Rules

The data protection rules are created using the advanced data privacy masking features of Watson Knowledge Catalog.

Create the Protect Credit Card Expiration Dates Rule

1. Select Add rule > New rule.

2. Select Data protection rule.

3. Click Next.

4. Enter Name = Protect Credit Card Expiration Dates

5. Enter Business definition = Protect all credit card expiration dates using the data privacy advanced masking method. (including the period at the end).

6. Build the rule as follows:

   • Criteria = If Data class contains any Credit Card Expiration Date.

   • Action = then mask data in columns containing Data class of Credit Card Expiration Date.

   • Select Obfuscate in the Select how to mask data: section.

   • Check Enable advanced masking options.

   • Select Identifier method as the Obfuscate method.

   • Select Irreversible masking for the Reversibility option.

   • Select Repeatable for the Consistency option.

   • Select No validation for the Input validation option.

   • Turn On the Auto refresh preview.

7. Click Create.

8. Select the Rules bread crumb, in top left corner to return to Rules main page.

Create the Protect Credit Card Numbers Rule

1. Select Add rule > New rule.

2. Select Data protection rule.

3. Click Next.

4. Enter Name = Protect Credit Card Numbers

5. Enter Business definition = Protect all credit card numbers using the data privacy advanced masking method. (including the period at the end).

6. Build the rule as follows:

   • Criteria = If Data class contains any Credit Card Number.

   • Action = then mask data in columns containing Data class of Credit Card Number.

   • Select Obfuscate in the Select how to mask data: section.

   • Check Enable advanced masking options.

   • Select Preserve format as the Obfuscate method.

   • Select Irreversible masking for the Reversibility option.

   • Select Repeatable for the Consistency option.

   • Select Input validation for the Input validation option.

   • Turn On the Auto refresh preview.

7. Click Create.

8. Select the Rules bread crumb, in top left corner to return to Rules main page.

Create the Protect Credit Card Validation Numbers Rule

1. Select Add rule > New rule.

2. Select Data protection rule.

3. Click Next.

4. Enter Name = Protect Credit Card Validation Numbers

5. Enter Business definition = Protect all credit card validation numbers using the data privacy advanced masking method. (including the period at the end).

6. Build the rule as follows:

   • Criteria = If Data class contains any Credit Card Validation Number.

   • Action = then mask data in columns containing Data class of Credit Card Validation Number.

   • Select Obfuscate in the Select how to mask data: section.

   • Check Enable advanced masking options.

   • Select Identifier method as the Obfuscate method.

   • Select Irreversible masking for the Reversibility option.

   • Select Repeatable for the Consistency option.

   • Select No validation for the Input validation option.

   • Turn On the Auto refresh preview.

7. Click Create.

8. Select the Rules bread crumb, in top left corner to return to Rules main page.

Create the Protect Email Addresses Rule

1. Select Add rule > New rule.

2. Select Data protection rule.

3. Click Next.

4. Enter Name = Protect Email Addresses

5. Enter Business definition = Protect all email addresses using the data privacy advanced masking method. (including the period at the end).

6. Build the rule as follows:

   • Criteria = If Data class contains any Email Address.

   • Action = then mask data in columns containing Data class of US Phone Number.

   • Select Obfuscate in the Select how to mask data: section.

   • Check the box to Enable advanced masking options.

   • Select Preserve format as the Obfuscate method.

   • Select Generate user name for the User name format option.

   • Select Custom for the Domain name option.

   • Enter outcomes.com for the Custom value.

   • Select Irreversible masking for the Reversibility option.

   • Select Repeatable for the Consistency option.

   • Select Input validation for the Input validation option.

   • Turn On the Auto refresh preview.

7. Click Create.

8. Select the Rules bread crumb, in top left corner to return to Rules main page.

Create the Protect Phone Numbers Rule

1. Select Add rule > New rule.

2. Select Data protection rule.

3. Click Next.

4. Enter Name = Protect International Phone Numbers

5. Enter Business definition = Protect all international phone numbers using the redaction masking method. There are no obfuscation masking options for the Phone Number data class. Obfuscation can only be applied to US Phone numbers. (including the period at the end).

6. Build the rule as follows:

   • Criteria = If Data class contains any Phone Number.

   • Action = then mask data in columns containing Data class of US Phone Number.

   • Select Redact in the Select how to mask data: section.

7. Click Create.

8. Select the Rules bread crumb, in top left corner to return to Rules main page.

Create the Protect US Phone Numbers Rule

1. Select Add rule > New rule.

2. Select Data protection rule.

3. Click Next.

4. Enter Name = Protect US Phone Numbers

5. Enter Business definition = Protect all US phone numbers using the data privacy advanced masking method. (including the period at the end).

6. Build the rule as follows:

   • Criteria = If Data class contains any US Phone Number.

   • Action = then mask data in columns containing Data class of US Phone Number.

   • Select Obfuscate in the Select how to mask data: section.

   • Check Enable advanced masking options.

   • Select Preserve format as the Obfuscate method.

   • Select Irreversible masking for the Reversibility option.

   • Select Repeatable for the Consistency option.

   • Select No validation for the Input validation option.

   • Turn On the Auto refresh preview.

7. Click Create.

8. Select the Rules bread crumb, in top left corner to return to Rules main page.

Create the Protect US Social Security Numbers Rule

1. Select Add rule > New rule.

2. Select Data protection rule.

3. Click Next.

4. Enter Name = Protect US Social Security Numbers

5. Enter Business definition = Protect all US social security numbers using the data privacy advanced masking method. (including the period at the end).

6. Build the rule as follows:

   • Criteria = If Data class contains any US Social Security Number.

   • Action = then mask data in columns containing Data class of US Social Security Number.

   • Select Obfuscate in the Select how to mask data: section.

   • Check Enable advanced masking options.

   • Select Preserve format as the Obfuscate method.

   • Select Irreversible masking for the Reversibility option.

   • Select Repeatable for the Consistency option.

   • Select No validation for the Input validation option.

   • Turn On the Auto refresh preview.

7. Click Create.

8. Select the Rules bread crumb, in top left corner to return to Rules main page.

8. Create Policies

The policies are created last because they have data protection and governance rules assigned to them. Therefore the rules have to be created first before the policy so that the rules can be assigned to the appropriate policies. For Outcomes, there is only one policy.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Governance > Policies.

3. Select Add policy > Import from file.

4. Click Add file.

5. Select the governance-policies.csv file from the download location.

6. Click Open.

7. Click Next.

8. Select Replace all values.

9. Click Import.

10. Click Go to Task.

11. Click Publish.

12. Go to the Cloud Pak for Data navigation menu.

13. Select Governance > Polices.

You should see 1 published policy named Protection of Sensitive Personal Information.

Assign Data Protection Rules to the Policy

1. Click the Protection of Sensitive Personal Information policy.

2. Scroll down to the Data protection rules section.

3. Click Add data protection rules.

4. Select the Protect Credit Card Expiration Dates rule.

5. Select the Protect Credit Card Numbers rule.

6. Select the Protect Credit Card Validation Numbers rule.

7. Select the Protect Email Addresses rule.

8. Select the Protect Phone Numbers rule.

9. Select the Protect US Social Security Numbers rule.

10. Click Add.

11. Click Publish.

12. Click Publish.

Create Platform Connections

1. Go to the Cloud Pak for Data navigation menu.

2. Select Data > Platform connections.

1. Create Amazon Object Store Connection

1. Select New connection.

2. Type amazon in the search area.

3. Click the Amazon S3 connector.

4. Click Select.

5. Enter the following properties:

• Name = Amazon Object Storage

• Description = Amazon S3 Object Storage bucket that contains data files used for analytics and AI. (including the period at the end)

• Bucket = Your Bucket

• Endpoint URL = Your URL

• Region = Your Regsion

• Authentication method = Basic credentials

• Access key = Your Access Key

• Secret key = Your Secret Key

6. Click Create.

2. Create Cloud Object Store Connection

1. Select New connection.

2. Type cloud in the search area.

3. Click the Cloud Object Storage connector.

4. Click Select.

5. Enter the following properties:

• Name = Cloud Object Storage

• Description = IBM Cloud Object Storage bucket that contains data files used for analytics and AI. (including the period at the end)

• Bucket = Your bucket

• Login URL = Your Login URL

• Authentication method = Resource Instance Id, API Key, Access Key and Secret Key

• Resource instance Id = Your resource instance Id

• API key = Your API Key

• Access Key = Your Access Key

• Secret Key = Your Secret Key

Note - The Resource Instance ID and API Key for the credential values are extracted from the JSON below:

CPD-Outcomes-Storage-Content-Reader { Your JSON }

6. Click Create.

3. Create Data Warehouse Connection

1. Click New connection.

2. Type db2 in the search area.

3. Click the Db2 Warehouse connector.

4. Click Select.

5. Enter the following properties:

• Name = Data Warehouse

• Description = Database that contains enterprise data needed by the business for analytics and AI. (including the period at the end)

• Database = BLUDB

• Host = Your Host

• Port = 50001

• Authentication Method = Username and password

• Username = Your Username

• Password = Your Password

• Certificates = Port is SSL-enabled

6. Click Create.

4. Create Document Store Connection

1. Select New connection.

2. Type mongo in the search area.

3. Click the MongoDB connector.

4. Click Select.

5. Enter the following properties:

• Name = Document Store

• Description = Data store that contains documents needed by the business for analytics and AI. (including the period at the end)

• Database = DOCUMENT

• Hostname = Your Hostname

• Port = Your Port

• Authentication database = admin

• Username = Your Username

• Password = Your Password

• Certificates = Port is SSL-enabled

5. Click Create.

5. Create Third Party Data Connection

1. Select New connection.

2. Type post in the search area.

3. Click the PostgreSQL connector.

4. Click Select.

5. Enter the following properties:

• Name = Third Party Data

• Description = Database that contains third party data needed by the business for analytics and AI. (including the period at the end)

• Database = Your Database

• Hostname = Your Hostname

• Port = Your Port

• Username = Your Username

• Password = Your Password

• Credentials = Port is SSL-enabled

6. Click Create.

Setup Data Virtualization

1. Enable Data Governance

This section enables the data governance capabilities in Data Virtualization to protect sensitive personal information and to publish virtualized tables to a governed catalog, the Governance catalog.

Select Settings > Service Settings from the Data Virtualization dropdown menu.

1. Select the Governance tab.

2. Click the Enforce policies within Data Virtualization to turn it on.

3. Click the Enforce publishing to a governed catalog to turn it on.

4. Select the Governance catalog.

2. Create Data Sources

Add the Data Virtualization data sources by using the connections that were just created in Platform connections by using the create from existing method.

1. Select Virtualization > Data Sources from the Data Virtualization dropdown menu.

2. Click Add connection.

3. Click Existing connection.

4. Select Third Party Data connection.

5. Click Add.

6. Click Add connection.

7. Click Existing connection.

8. Select Document Store connection.

9. Click Add.

10. Click Add connection.

11. Click Existing connection.

12. Select Data Warehouse connection.

13. Click Add.

14. Click Add connection.

15. Click Existing connection.

16. Select Cloud Object Storage connection.

17. Click Add.

18. Click Add connection.

19. Click Existing connection.

20. Select Amazon Object Storage connection.

21. Click Add.

3. Virtualize the Data

We only virtualize the tables needed for analytics and AI modeling tasks that will be used to create virtual views or that are added to projects so data transformation and preparation can be performed prior to doing analytics and AI tasks. For the Outcomes data fabric demonstrations, 12 tables will be virtualized.

1. Select Virtualization > Virtualize from the Data Virtualization dropdown menu.

2. Click the Search asset type filter dropdown menu.

3. Select Table.

4. Select the Settings menu icon on far right next to the refresh icon.

5. Uncheck Hostname:port and Columns.

Virtualize Banking Tables

1. Sort by Table in ascending order (Arrow pointing up).

2. Type mortgage in search area.

3. Select the checkbox next to the following Banking tables:

   • MORTGAGE_APPLICANT

   • MORTGAGE_APPLICATION

   • MORTGAGE_CANDIDATE

4. Click Add to cart.

5. Click the x at the end of the search area to clear it out.

Virtualize Employee Tables

1. Sort by Table in ascending order (Arrow pointing up).

2. Type employee in search area.

3. Select the following Employee tables:

   • EMPLOYEE_HISTORY

   • EMPLOYEE

   • EMPLOYEE_EXPENSE_DETAIL

   • EMPLOYEE_SUMMARY

4. Click Add to cart.

5. Click the x at the end of the search area to clear it out.

6. Type position in the search area.

7. Select the checkbox next to the POSITION_DEPARTMENT table.

8. Click Add to cart.

9. Click the x at the end of the search area to clear it out.

10. Type ranking in the search area.

11. Select the checkbox next to the RANKING_RESULTS table.

12. Click Add to cart.

13. Click the x at the end of the search area to clear it out.

14. Type training in the search area.

15. Select the checkbox next to the TRAINING_DETAILS table.

16. Click Add to cart.

17. Click the x at the end of the search area to clear it out.

Virtualize Customer Tables

1. Type customer in search area.

2. Select the checkbox next to the following Customer tables:

   • CUSTOMER

   • CUSTOMER_LOYALTY

3. Click Add to cart.

4. Click View cart (12).

Virtualize and Publish

Before we virtualize the tables and publish them to the Governance catalog we need to make a few changes to the virtualized table schema names instead of using the not so user friendly schema names that are automatically generated by the Watson Query service.

1. Click the Virtutalized data radio button in the Assign to section.

Note - The Publish to section has automatically checked Catalog to publish to and has the Governance catalog automatically selected. This is due to the governance settings we set in the previous section.

1. Change Banking Tables Schema

1. Go to the first table in the list with a Source schema of BANKING.

2. Click the X to clear out the default Schema assigned.

3. Type BANKING (in uppercase) in the Schema name area and select Create BANKING.

4. Repeat the following steps for all the tables that have a Source schema of BANKING:

   • Click the X to clear out the default Schema assigned.

   • Select the drop down arrow v in the Schema name area.

   • Select the BANKING schema from the list of schema names.

2. Change Employee Tables Schema

1. Go to the first table in the list with a Source schema of EMPLOYEE.

2. Click the X to clear out the default Schema assigned.

3. Type EMPLOYEE (in uppercase) in the Schema name area and select Create EMPLOYEE.

4. Repeat the following steps for all the tables that have a Source schema of EMPLOYEE:

   • Click the X to clear out the default Schema assigned.

   • Select the drop down arrow v in the Schema name area.

   • Select the EMPLOYEE schema from the list of schema names.

3. Change Customer Tables Schema

1. Go to the first table in the list with a Source schema of CUSTOMER.

2. Click the X to clear out the default Schema assigned.

3. Type CUSTOMER (in uppercase) in the Schema name area and select Create CUSTOMER.

4. Repeat the following steps for all the tables that have a Source schema of CUSTOMER:

   • Click the X to clear out the default Schema assigned.

   • Select the drop down arrow v in the Schema name area.

   • Select the CUSTOMER schema from the list of schema names.

4. Virtualize the Tables

Note - Before you select the button to Virtualize the tables, make sure that all of the tables have the new schema names you just assigned!

Watson Query has a mind of its own. Sometimes, for whatever reason, when you change the schema names, as I outline in the next section, Watson Query makes it look like the schema name has been changed, but when you select the Virtualize button, it is possible that one of the tables still has the old funky schema name assigned to it. Unfortunately, If that happens, there is no turning back or cancellation button. Therefore, you have to delete all the virtual tables one by one (no bulk delete option either...) and then go back to Step 3 of this section and start the process all over again from selection to virtualization. It seems overkill, but trust me, it is important to our demo that it all be displayed correctly.

1. Click Virtualize.

2. Click Confirm but do not check the Do not show this message again checkbox.

   • Wait for all tables to be virtualized and published before proceeding to the next step. The Virtualization Status and Publish Status indicators for all 12 tables should have a green check mark next to them.

   • Also make sure that your schema changes are correct by scrolling down the list of tables and verifying that you have 3 tables with the BANKING schema, 2 tables with the CUSTOMER schema and 7 tables with the EMPLOYEE schema.

3. Click View virtualized data.

There should be 12 virtualized tables in the Virtualized data section.

5. Grant User Role to Tables

1. Click the Table column header and sort the tables in ascending order.

2. Go to the first table in the list of virtualized tables.

   • Based on my experience on how Data Virtualization behaves, it should be the CUSTOMER or CUSTOMER_LOYALTY table since they were the last ones virtualized.

3. Click the ellipses... next to the table.

4. Click Manage access.

5. Select Specific user.

   • It should be selected as the default.

6. Select the Roles tab.

7. Click Grant access.

8. Select the User role.

9. Click Add roles.

10. Select the Back button.

Repeat steps 2-10 above for the remaining 11 virtualized tables.

When you get to the 10th table in the list), select the Items per page control and click on 20 for the remaining 2 tables to show up in the list.

To verify you did not miss granting access to any tables:

1. Select User management from the Data Virtualization dropdown menu.

2. Click the Roles tab.

Make sure the User role has the number 12 listed in the Granted access column.

Stay in the User Management section for the next task.a

5. Add CPD Business User Access

From the User Management section:

1. Click the Users tab.

2. Click Grant access.

3. Type in an email address = [email protected]

4. Click the User role radio button.

5. Click Create.

Note - It will take a couple of minutes (usually 1-2) for the user access to be assigned so be patient and wait until it completes!

Validate Published Data Assets

We published the virtualized tables to the Governance catalog during the virtualization process so that any of the tables that contain sensitive information will be autonomously protected by the data protection rules defined. Data virtualization does some things automatically during publishing that need to be modified and verified. We will perform those steps in the following sections.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Catalogs > All catalog.

3. Select the Governance catalog.

1. Update Data Virtualization Connection

Data virtualization automatically creates a connection to the Watson Query service in the Governance catalog when it publishes data assets to a governed catalog. However, the connection name is not very user friendly and is owned by a user named unavailable. We will change the name, description and owner in the following steps:

1. Click the Filter by dropdown.

2. Select Connections.

3. Click on the only connection in the data asset list.

There should be one connection in the Governance catalog that is named Data Virtualization_ea660a45-a34d-4961-afb9-39875a62e7b4 The connection name is always named "Data Virtualization" with the "Instance Id" (the value after the plus sign) of the Watson Query service it is associated with appended to the end. The instance id will be different for production and test.

4. Click the Asset tab.

5. Click here to edit the connection details.

6. Clear out the Name.

7. Enter a new Name = Watson Query.

8. Enter a Description = Connection to the Watson Query data virtualization service in the Cloud Pak for Data Outcomes resource group in this cloud account. (including the period at the end).

9. Click Credentials from the left side menu.

10. Select the Authentication Method and choose API Key.

11. Enter an API Key of:

    Prod = UcBbBseZBsuE6ITZAHU5Suo7yf8LdwhkaKk22lHkxz2V

    Test = syQkhcXhJMkvqrMQtfYZzU1y7Nf9rxirEcAmrYJovEBB

12. Check the box Port is SSL-enabled.

13. Click the Test connection button.

    The test should succeed. If not, go back and re-check all the entries to make sure you copied and pasted them correctly. When it does succeed you can proceed to the next step to create it.

14. Click Save.

15. Click the Governance cookie in the menu to get back to the main page of the catalog.

2. Validate Data Profiles

We need to ensure the 12 tables that were published have been profiled. This is especially important for the 3 tables that contain sensitive data; MORTGAGE_APPLICANT, CUSTOMER and EMPLOYEE. They have to have all their columns classified correctly because the data protection rules are triggered based on data class. If the columns that contain sensitive data are not classified correctly, like a US Social Security Number or Credit Card Number, the data protection rules will not recognize them as such and therefore they will not be masked.

From the Assets section of the catalog, verify that there are 12 tables. They should be sorted in alphabetical order by "schema.table name". Starting from the first table in the list, do the following:

1. Click on the Name of the first data asset in the list.

Note - The first table should be the BANKING.MORTGAGE_APPLICANT table.

2. Click on the Profile tab.

Note - You should see a profile created for every data asset. If not, click the Create profile button. The profile will take a while to create in the background so you do not have to wait for it to complete. Create the profile and move on to the next data asset in the list. If profiles continue to fail to be created, open up a support ticket because there is an issue with the Watson Knowledge Catalog data profiling job.

3. Click the Governance cookie in the menu to get back to the main page of the catalog to select the next table in the list.

4. Repeat steps 1-3 above for the remaining 11 tables.

3. Validate Sensitive Data Classes

From the data asset list we need to go back to the profile for the 3 tables that contain sensitive information and check their column classifications.

MORTGAGE_APPLICANT

1. Click on the BANKING.MORTGAGE_APPLICANT data asset.

2. Click on the Profile tab.

3. Scroll to the right.

4. Click the classification dropdown on the EDUCATION column.

5. Select View all.

6. Type in edu.

7. Select Education Status.

8. Click Add.

9. Click the classification dropdown on the EMPLOYMENT_STATUS column.

10. Select View all.

11. Type in emp.

12. Select Employment Status.

13. Click Add.

Click the Governance cookie in the menu to get back to the main page of the catalog to process the next table.

CUSTOMER

1. Click on the CUSTOMER data asset.

2. Click on the Profile tab.

3. Scroll to the right.

4. Click the classification dropdown on the STATE_CODE column.

5. Select View all.

6. Type in state/.

7. Select State/Providence Code.

8. Click Add.

9. Click the classification dropdown on the POSTAL_CODE column.

10. Select View all.

11. Type in postal.

12. Select Postal Code.

13. Click Add.

14. Click the classification dropdown on the EDUCATION column.

15. Select View all.

16. Type in edu.

17. Select Education Status.

18. Click Add.

19. Click the classification dropdown on the LOCATION_CODE column.

20. Select View all.

21. Type in code.

22. Select Code.

23. Click Add.

24. Click the classification dropdown on the INCOME column.

25. Select View all.

26. Type in income.

27. Select Income.

28. Click Add.

29. Click the classification dropdown on the CREDIT_CARD_TYPE column.

30. Select View all.

31. Type in credit.

32. Select Credit Card Network.

33. Click Add.

34. Click the classification dropdown on the CREDIT_CARD_CVV column.

35. Select View all.

36. Type in credit.

37. Select Credit Card Validation Number.

38. Click Add.

39. Click the classification dropdown on the CREDIT_CARD_EXPIRY column.

40. Select View all.

41. Type in credit.

42. Select Credit Card Expiration Date.

43. Click Add.

Click the Governance cookie in the menu to get back to the main page of the catalog to process the next table.

EMPLOYEE

1. Click on the EMPLOYEE data asset.

2. Click on the Profile tab.

3. Click the classification dropdown on the EMPLOYEE_CODE column.

4. Select Identifier.

5. Click the classification dropdown on the FIRST_NAME_MB column.

6. Select First Name.

7. Click the classification dropdown on the GENDER_CODE column.

8. Select View all.

9. Type in gender.

10. Select Gender.

11. Click Add.

12. Click the classification dropdown on the WORK_PHONE column.

13. Select View all.

14. Type in phone.

15. Select Phone Number.

16. Click Add.

17. Click the classification dropdown on the EXTENSION column.

18. Select View all.

19. Type in text.

20. Select Text.

21. Click Add.

22. Click the classification dropdown on the FAX column.

23. Select View all.

24. Type in phone.

25. Select Phone Number.

26. Click Add.

27. Click the classification dropdown on the COMMUTE_TIME column.

28. Select View all.

29. Type in quantity.

30. Select Quantity.

31. Click Add.

Configure Watson OpenScale and Studio

1. Create Outcomes Runbook Project

Everything is now in place to build out the Business catalog with the proper connections, data assets, associated metadata from the business glossary, and data protection rules in place to protect sensitive and personally identifiable information. We will build out the catalog using a project that was populated using metadata import and metadata enrichment features.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Projects > view all projects.

3. Click New project +.

4. Click Create a project from a sample file.

5. Click browse to upload the project file.

6. Select the Outcomes-Runbook-Project-CPDaaS.zip file from your download location.

7. Enter a Name = Outcomes Runbook.

8. Enter a Description = Administrative project used to configure and manage Cloud Pak for Data Outcomes. (including the period at the end).

9. Click Create.

You will see a dialog box with the message that the Outcomes Runbook project is being created....

10. Click View new project.

The project should create successfully and have 10 assets.

Populate Business Catalog

Data Warehouse - Import the cloud object storage files that are published to the Business catalog.

1. Create Outcomes Catalog Project

Everything is now in place to build out the Business catalog with the proper connections, data assets, associated metadata from the business glossary, and data protection rules in place to protect sensitive and personally identifiable information. We will build out the catalog using a project that was populated using metadata import and metadata enrichment features.

1. Go to the Cloud Pak for Data navigation menu.

2. Select Projects > view all projects.

3. Click New project +.

4. Click Create a project from a sample file.

5. Click browse to upload the project file.

6. Select the Outcomes-Catalog-Project-CPDaaS.zip file from your download location.

7. Enter a Name = Outcomes Catalog.

8. Enter a Description = Administrative project used to configure the Business catalog for Cloud Pak for Data Outcomes. (including the period at the end).

9. Click Create.

You will see a dialog box with the message that the Business Catalog project is being created....

10. Click View new project.

The project should create successfully and have 47 data assets.

Connection Metadata

Each connection in the project has been assigned a name, description, and tags so they will be carried over into the Business catalog when they are published. The table below contains this metadata in case of lose. This is for reference only.

Data Asset Metadata

Each data asset in the project has been assigned a name, description, and tags so they will be carried over into the Business catalog when they are published. The table below contains this metadata in case of lose. This is for reference only.

Publish Assets to Business Catalog

We will now publish the 47 data assets from the project to the Business Catalog. Every asset in the project should have the metadata outline in the tables above already applied; tags and descriptions.

Note - Watson Knowledge Catalog has a strange, and frankly speaking, random means of publishing bulk items form a project to a catalog. They do not always appear in the order you have chosen to publish.

The assets will be published in a specific order, as stated below, so that the Recently Added category in the catalog gets populated properly with the tables that are searched for, and that are used in the analytics project, appearing as the most recent tables added to the catalog. This helps during the demo so end users can see and find the tables we want them too fast and easily.

1. Publish Connections

We will publish the data connections first, in the order they are listed in the instructions below. This will put them at the very bottom of the recently added list and in the order they are used in our demos and not interfere with the primary data assets we want to appear first in that category.

Note - Tags are case sensitive so all the tags need to be entered in mixed case as instructed. This is to ensure consistency as they appear in the catalog and there are not duplicate tags created in the tags selection dropdown in the catalog that are the same but in different cases.

From the Assets tab using the Asset type filter on the left:

1. Click Data access.

2. Click Connection.

Publish Amazon Object Storage

1. Click the checkbox next to Amazon Object Storage.

2. Click Publish to catalog.

3. Select the Business catalog as the Target.

4. Click in the tags area.

5. Enter Employee and click the + sign.

6. Enter Warehouse and click the + sign.

7. Click Publish.

Publish Cloud Object Storage

1. Uncheck the checkbox next to Amazon Object Storage.

2. Click the checkbox next to Cloud Object Storage.

3. Click Publish to catalog.

4. Select the Business catalog as the Target.

5. Click in the tags area.

6. Enter Employee and click the + sign.

7. Enter Warehouse and click the + sign.

8. Click Publish.

Publish Document Store

1. Uncheck the checkbox next to Cloud Object Storage.

2. Click the checkbox next to Document Store.

3. Click Publish to catalog.

4. Select the Business catalog as the Target.

5. Click in the tags area.

6. Enter Employee and click the + sign.

7. Click Publish.

Publish Third Party Data

1. Uncheck the checkbox next to Document Store.

2. Click the checkbox next to Third Party Data.

3. Click Publish to catalog.

4. Select the Business catalog as the Target.

5. Click in the tags area.

6. Enter Banking and click the + sign.

7. Enter Customer and click the + sign.

8. Enter Employee and click the + sign.

9. Enter Mortgage and click the + sign.

10. Click Publish.