Getting started

About this task

Use IBM Data Virtualization Manager for z/OS (Data Virtualization Manager) to securely connect to and access data from your z/OS® mainframe environment.

This section provides an overview of the general tasks that are associated with implementing Data Virtualization Manager within your existing z/OS mainframe environment to get virtual access to your mainframe data from your applications.

The following illustration shows the key Data Virtualization Manager components within the Data Virtualization Manager architecture.

Task 1: Install and configure the Data Virtualization Manager server

About this task

Install the Data Virtualization Manager server using IBM SMP/E for z/OS.

Procedure

Use the information in Program Directory for IBM Data Virtualization Manager for z/OS to install the Data Virtualization Manager server on your system.

What to do next

Install the drivers.

Task 2: Install the drivers

About this task

Install the drivers on your development workstation that are required to access data made available through Data Virtualization Manager. Data Virtualization Manager supports the following drivers:

JDBC
ODBC

Procedure

See the IBM Data Virtualization Manager Installation and Customization Guide for details about installing the drivers.

What to do next

Install and configure the Data Virtualization Manager studio.

Task 3: Install and configure the Data Virtualization Manager Studio

About this task

Install and configure the Data Virtualization Manager Studio on your Windows system.

Procedure

See the IBM Data Virtualization Manager Installation and Customization Guide for details about installing and configuring the IBM Data Virtualization Manager studio.

What to do next

Use the Data Virtualization Manager studio to virtualize the mainframe data.

Task 4: Virtualize your mainframe data

About this task

To use SQL or NoSQL to access your mainframe data, use the Data Virtualization Manager studio to create the metadata on the Data Virtualization Manager server. The metadata is used to reference source libraries that already exist on your mainframe. The information that is required to virtualize the source data depends on the data type. For example, for a VSAM file, the virtual source library must contain the copybook that describes the structure of the records in the VSAM file. For an IMS database, you need to provide the Database Definition (DBD) and Program Specification Block (PSB) files, and a copybook structure for each segment of the IMS database that you want to virtualize.

Note: For complete task details, see Creating server metadata.

Procedure

Use the Studio to connect to the Data Virtualization Manager server.
Use the Create Virtual Source Library wizard to create virtual source libraries that reference the data layouts on the mainframe.
Create a virtual table (virtual representation) that maps to the data that you want to access:
- For SQL access to data, complete the virtual table wizard that is appropriate for the data that you want to access.
- For NoSQL access to data, complete the wizard that is appropriate for the data that you want to access.

What to do next

Generate and execute the SQL that is used to test access to the data on the mainframe and to review the resulting data.

Task 5: Generate and execute a query

About this task

To test access to your data, generate and execute a query from an existing virtual table.

Procedure

On the Studio Server tab, right-click the virtual table that maps to the data that you want to access and select Generate Query.
Depending on your preference settings, the query is automatically executed after being generated or you are prompted to execute the statement. If prompted, click Yes.
In the SQL Results view, review the result set that displays. If necessary, in the Generated.sql view, modify the SQL to select only the data that you want to access. Any ANSI compliant SQL is acceptable.

What to do next

You can now generate the code from the SQL.

Task 6: Generate sample code

About this task

To access mainframe data from your programs and applications, use the Studio to generate sample code.

Note: For task details, see the Generating code from SQL.

Generate sample code for SQL access to your data

Procedure

Right-click the virtual table or the selected SQL statement in the Generated.sql view and select Generate Code From SQL.
Complete the Code wizard to generate the code for the programming language/environment that you want to use. Choices include:
- Java JDBC Class
- Java Spark Application
- Scala Jupyter Notebook
  Note: The generated Scala Jupyter Notebook code provides a simple starter program showing you how to use the JDBC driver to load data into a Spark application. The wizard does not allow SQL parameter markers for this application type.
- Scala Spark Application
- VB.NET Class
- C# Class
- Other (requires that you specify XSLT file and file extension)

Generate sample code for NoSQL access to your data

Procedure

Select the virtual collection that represents the data you want to access.
Select Generate Query and select either find(), find(...), or count().
In the *generated.js view, right-click the selected query and select Execute Mongo Script.

Results

Use the generated sample code in your programs and applications.