Important: IBM Cloud Pak® for Data
Version 4.6 will reach end of support (EOS) on 31 July, 2025. For more information, see the
Discontinuance of service announcement for
IBM Cloud Pak for Data Version 4.X.
Upgrade to IBM Software Hub Version
5.1 before IBM Cloud Pak for Data Version 4.6 reaches end of
support. For more information, see Upgrading IBM Software Hub in the IBM Software Hub Version 5.1
documentation.
You can join multiple tables from multiple data sources into a single virtual table,
which is also known as a join view.
Remember:
The data requests
() feature
is deprecated and will be removed in a future release. Consider
workflows instead.
Procedure
To create a virtual view from existing virtualized tables, complete the following steps.
- On the navigation menu, click .
The service menu opens to the Data
sources page by default.
- On the service menu, click
.
Your existing virtualized tables are listed.
- Select two tables that you want to join and click Join to display
the Join virtual objects window.
Tip: If you prefer, you can click to skip the following steps and use the IBM common SQL engine
instead. See the
SQL Reference for more details on SQL syntax and function
compatibility.
- Use the graphical join wizard to select at least one join key, which
consists of a pair of columns, of the same data type, from both of the virtualized tables. Then,
from each table, select the columns to include in the join results. This step does not copy or move
any data. It creates a table definition that is a combination of the two tables.
To create a join key by using only the keyboard, follow these steps:
- Press Enter to select a column name in a row in Table 1.
- Press the Tab key to navigate to a column name in a row in Table 2.
- Press Enter on the row in Table 2. The rows are joined.
Restrictions:
- If you are joining tables with many rows, the preview of the join might time out after
approximately 10 minutes if the data sources are unable to complete the processing of the join.
- If the columns of the tables that you are joining do not share any common data,
the preview of your join view might be empty. You can continue to join your virtual objects;
however, the join view might not contain valuable data. Any data that you add to the virtual objects
is automatically reflected in the join view.
- You can join only two tables at one time. To join more than two tables, join two tables and get
a view. Then, join the view and the third table to get another view.
-
Privileges and authorities that are granted to user groups are not
considered when you create views. This limitation is a result of a Db2® limitation on groups.
For more information, see Privileges and authorities that are
granted to user groups are not considered when you create views.
- Click Next. You can use the new table to query the data from both
of the base tables.
- On the Edit column names page, enter a View
name, select a Schema, and edit column names as needed.
- Select where to assign the view with one of the following options:
| Assign to |
When to use this option |
| Data request |
Select Data request if you created the virtual table in response to a
data request. Then, choose the appropriate request. The table also appears in Virtualized
data. |
| Project |
Select Project if you created the virtual table to use in a specific
project. Then, choose the appropriate project. The table also appears in Virtualized
data. |
| Virtualized data |
Select Virtualized data if the table was not created in response to a data request or to use in a specific project. This setting is
the default if no data requests or projects exist. |
-
Click Create view to complete the process.
Results
If Watson
Query and Watson Knowledge
Catalog are installed in the same OpenShift® project (namespace), your virtual object is
published to the primary catalog.
What to do next
You can use the virtualized data in a number of different ways. For example,
you can use them in a Jupyter Notebook, create new models within the Model Builder, or build charts
or graphics on the analytics dashboard.