Virtualizing data

Virtualizing data (Data Virtualization)

Use Data Virtualization to create a virtual table to segment or combine data from one or more tables. Data Virtualization connects multiple data sources into a single self-balancing collection of data sources or databases.

Overview

This service is not available by default. You must create and deploy a service instance of Data Virtualization in Cloud Pak for Data as a Service to virtualize your data. For more information, see Provision a service instance for Data Virtualization.

Things to know about using Data Virtualization:

Data Virtualization has no required prerequisite services or service integrations.
Data Virtualization works with data. For more information, see Supported data sources (Data Virtualization as a service).
Data size limits are defined by the data source. For more information, see Data source considerations (Data Virtualization).
For known issues and limitations, see Known issues and limitations.
Data Virtualization uses your IBM Cloud credentials to connect to the service. You must have certain Data Virtualization roles to perform certain tasks. For more information, see User credentials and supported authentication methods.

The most common mechanism for virtualizing data is to create a table view or virtual table.

The image shows how Data Virtualization connects multiple data sources into a single self-balancing collection of data sources or databases. — Figure 1. Connection of multiple data sources into a single collection

Tables from multiple sources that are similar can be combined into a single virtual table, which creates a unified definition that contains the columns and data from all participating data sources. Segmentation is vertical (either a subset or superset of columns based on a selection of chosen columns). You can then run queries against the resulting virtual table no different than how you would query any of the base tables. These tables are referred to as grouped tables. For more information, see Creating a virtualized table from multiple data sources (Data Virtualization).

After you provision the Data Virtualization service, you can manage users, connect to multiple data sources, create and govern virtual assets, then consume the virtualized data.

Connect, Join, Create Views, and Consume are the main actions that are needed for Data Virtualization. — Figure 2. Virtualizing data with the Data Virtualization service

Connect: Start by connecting to data sources. You can connect to multiple data sources. For more information, see Adding data sources and Supported data sources (Data Virtualization as a service).
Join, create, and govern: Then, create virtual tables, group tables by schema, associate data with projects, and govern your virtual assets. For more information, see Creating a virtualized table and Governing virtual data (Data Virtualization).
Consume: Finally, consume virtual tables in analytics projects, dashboards, data catalogs, and other applications. For more information, see Dashboard services.

Learn more

For more information about these tasks, see the following resources.