Table of contents

Responding to data requests

You can respond to requests for data from other IBM® Cloud Pak for Data users.

Before you begin

If you decide that a request requires data to be virtualized, the Data Virtualization service must be installed and you must have the Data virtualization administrator role in Data Virtualization to fulfill the request.

About this task

When a user needs data but cannot find it in the catalog or does not have the appropriate permissions to access the data, the user can submit a data request. At a minimum, the request includes the name of the project for which the user needs the data and a description of the data that they need for the project. The user can also specify the types of data that they need as columns.

You can respond to the data request in multiple ways. Unless your organization has specific guidance on how to respond to data requests, there is no single right way to respond. When you respond to the data request, consider the following factors:
  • Can you find a single data set that contains the requested data in the enterprise data catalog?
  • Can you find data sets that contain only pieces of the requested data? For example, is all of the data available but spread across multiple tables?
  • Do you want to provide exactly the information that the user is looking for? For example, if the user requests a data set that contains three specific columns, do you want to provide a data set that contains only those three columns?
  • Do you want to provide the entire data set, even if it includes more information than the user requested?

You might not be able to find exactly what the user is requesting in exactly the right format. The request is intended to help you understand the type of information that the user is looking for.

If the data already exists in the catalog, but the user does not have access to the data, you can:
  • Associate a table that contains that information with the specified project and give the user credentials to access the table.
  • Virtualize the data to display only the requested columns in the resulting view.
If the data exists in multiple tables, you can:
  • Run a transformation job to join the tables and then create a connection to that data source.
  • Virtualize the tables so that the resulting view contains the requested columns (or optionally include other columns from the tables).

Procedure

  1. Click Data requests on your home page.
  2. Open the data request that you want to work on by clicking it.
  3. To begin work on the request, click Claim.
  4. Determine how to respond to the request.
    • If you are connecting to a data source, click the Source tab, and enter the appropriate information so that the user can connect to the data source.
    • If you are virtualizing data, in the drop-down menu, click Virtualize. See Virtualizing data (Data Virtualization) to learn how virtualize data with Data Virtualization.

      After you have virtualized the data for the request, click the Source tab, and enter the appropriate information so that the user can connect to the data source.

      If the request was submitted by a Data Virtualization user and you select Data Virtualization to fulfill it, the user's Data Virtualization username and password are automatically provided in the data request.

    • If you are transforming data, in the drop-down menu, click Transform. See Transforming data (DataStage®) to learn how to transform data with DataStage.

      After you have transformed the data for the request, click the Source tab, and enter the appropriate information so that the user can connect to the data source.

  5. When you have fulfilled the data request, click Deliver.

Results

The request is returned to the user who submitted the request. The user can then accept the request.