Virtualizing data

Use the Data Virtualization service to easily join data from different sources in one unified view, without manual changes, data movement, or replication.

The Data Virtualization service is not available by default. An administrator must install the service. To determine whether the service is installed, open the Services catalog. If the service is installed and ready to use, the tile in the catalog shows Ready to use.

Overview

With Data Virtualization, you can access physical data from multiple sources through a single semantic virtual layer, meaning that the data can be accessed, manipulated, and analyzed without the need to know its physical format or location, and without having to move or copy it.

Data Virtualization is part of the data fabric.

What's new in Data Virtualization

To view a list of the new Data Virtualization features, see New and enhanced features in Data Virtualization.

Prerequisites

Data Virtualization has no prerequisite services or service integrations. Data Virtualization provisions IBM Db2 Data Management Console if it is not provisioned already.

If you want to publish your virtual data to a governed catalog, you must install IBM® Knowledge Catalog. For more information, see Data governance (IBM Knowledge Catalog).

Getting started

To start using Data Virtualization, follow these high-level steps:
  1. Open the Data Virtualization service.
    In the IBM Software Hub navigation menu, select Data > Data virtualization.
    Screenshot of Cloud pak for Data drop down menu with Data Virtualization
  2. Add your data sources to Data Virtualization.
    Navigate to the Data sources page and then select Add connection to add connections. Data Virtualization supports dozens of relational and nonrelational data sources. Screenshot of Data sources page
    screenshot of New Connections page feature IBM, third-party and user-defined connection options
  3. Virtualize the tables from the data source.
    In the Virtualize page, select the tables you want to virtualize and then select Add to cart > View cart to virtualize the tables.
    Screenshot of virtualize page with objects selected
    Screenshot of Review cart and virtualize tables page
  4. Join the tables to create a unified view.
    In the Virtualized data page, select the tables you want to join and then select Join to join the objects.
    screenshot of Virtualized data page with objects selected to joinScreenshot of Join virtual objects page with objects being joined
  5. Query the virtual objects.
    Navigate to the Run SQL page to query your virtual objects using the built in SQL editor.
    Screenshot of Run SQL page
  6. Consume the data using other IBM Software Hub services in the data fabric.
  7. Consume virtual tables in projects, dashboards, data catalogs, and other applications.

Learn more

For more information on supported data sources, see Supported data sources in Data Virtualization.

For more information on known issues and limitations, see Known issues and limitations for Data Virtualization.