1. Create the server data sets through the hlq.SAVZCNTL members AVZDFDIV, AVZGNMP1 and AVZEXSW1
2. Set up the security application to use with the server by using one of the following hlq.SAVZCNTL members: AVZRAVDB, AVZA2VDB, AVZTSVD
3. Configure Workload Manager (WLM) for optimum performance of the server
4. Authorize the product LOAD library by using the Program Facility (APF)
5. Create a copy of the product libraries
6. Configure the server to support DB/CS (database/data communication) system to enable online use of databases
7. Customize the server to access your data sources in hlq.SAVZEXEC(AVZSIN00)
8. Configure the started task JCL located in hlq.SAVZCNTL(AVZ1PROC) before you can start the server
9. Configure the Command List (CLIST) that invokes ISPF panels by using hlq.SAVZEXEC(AVZ)
10. Verify the installation by creating a virtual table and accessing its underlying VSAM file

Create the DVM Server data sets

The first step is to create server data sets required for data processing. One data set is called a global variable file that simply retains parameters for its use. A server trace file can be used to record messages about server operations during processing.

Within the batch job called AVZDFDIV, you can change the product high-level qualifier and choose a forward character server subsystem name that's unique for this server. The second and third characters must be VZ, but the 1st and 4th characters can be anything you like. The following example is named AVZS.

The second batch job to run is called AVZGNMP1 to generate a user-defined map file. As a user creates maps or edits existing maps, they're saved in the user-defined map file instead of being stored in the map file supplied by the product data sets. Change HLQ1 to the high-level qualifier of the product data sets and change HLQ2 to a value to create the map file data set.

Change AVZS to a wanted server subsystem name. Anyone that uses this product requires UPDATE authority to this map data set so that maps can be created and then stored in the user-was defined map file.

Set up the Security Application

The second step is to define security authorizations for the server based on the security application in use at your data center. You might need to involve your z/OS® security specialist to ensure you proceed with the correct batch job. There are three batch jobs located in the SAVZCNTL data set.

Additionally, the started task user ID requires READ, EXECUTE, READ or READ/WRITE/UPDATE to the following data sets.

Run the appropriate job for your security application from within hlq.SAVZCNTL:

• AVZRAVDB - IBM Resource Access Control Facility (RACF®) security
• AVZA2VDB - CA AFC2 security
• AVZTSVDB - CA top secret security

Configure the Workload Manager

The third step involves configuring an appropriate WLM policy for the DVM Server.  The following guidance can be used for the installation and customization. System Programmers are familiar with the steps associated with WLM policy definitions.

A best practice is to configure the DVM Server to use a medium to high performing WLM velocity goal, as its default service class.

2. Create a WLM classification rule
   • From the WLM ISPF application, select option 6 (Classification Rules)
   • Select option '1' to create a rule
   • Set the Subsystem Type to AVZ and provide an optional description
   • When a default service class name doesn't exist, select option '4' (Service Classes) from the WLM menu and press enter and PF3 to save ​​​​​​
3. Define the DVM Server started task AVZ1PROC to a WLM service class
   • From the WLM ISPF application, select option '6' (Classification Rules)
   • For STC WLM-subsystem type, select Modify
   • Add an entry for AVZ1PROC
   • Provide an appropriate service class for the started task and define it relative to existing workload management objectives
   • Add a unique Report class for the started task
4. Activate the new WLM Policy definition

Authorize the program LOAD Library

The load library needs to be API-Authorized by using the following command:

• Authorize hlq.SAVZLOAD

Create a backup of Product libraries

Create a copy of the product libraries generated from the SMP installation as a best practice and creates a backup for recovery, when any customization needs to be retained. When you don’t create a copy of your product data sets, any SMP maintenance you apply going forward overlays any customizations made.

• Copy the product libraries to a runtime set of libraries
• IEFBR14/IEBCOPY job
• ADRDSSU DUMP job

Configure support for the DB/CS system

Configure the initialization Member, also known as the Server Configuration Member in the exec data set.

The name of the member depends upon the four characters Subsystem ID you assigned the server. Therefore, we called it an AVZS, so the initialization member is called AVZSIN00.

• HLQ.SAVZEXEC(AVZAIN00) is used to configure support for Japanese code pages and double-byte sets

• SQLENGDFLTCCSID is the default CCSID to use when processing data via the server for non-Db2 data
• SQLENGDBCSLTFMT is the default DBCS format to use when translating a single-byte code page to a double-byte code page.  Letters on single-byte code pages are translated to full-width DBCS characters.

Customize the DVM Server for access to databases

Customize the HLQ values in the initialization member to point to those libraries that you created the product with. HLQ1 is for the product target libraries that you created during the SMP installation process. HLQ2 represents the runtime libraries for the product. When runtime libraries are not created from Step 5, map HLQ2 to the product data sets that were created during the SMP installation process.

• HLQ.SAVZEXEC(AVZSIN00)
• Provide the high-level qualifier of the installation libraries in SHLQ1
• Provide the high-level qualifier of the runtime libraries in SHLQ2
• When there are no runtime libraries, then SHLQ2 and SHLQ1are recommended having the same value

During this step, you can customize the server initialization member for the TCP/IP ports that the server needs to communicate.  You need to have a dedicated web service port number (WSOEPORT) that the Eclipse plug-in can use to communicate with the DVM Server.

Relational database customization
Following are examples of how to set up the initialization member to access relational database management systems. However, some data sources don't need any customization, like VSAM. Data sources like Db2 distributed databases, Oracle, Teradata, or Amazon Redshift do require customization.

In our example, there are three different databases configured. Notice there is a pattern for each defined database block with subtle changes that require database administrators for each data source to get specific values for the unique parameters for each.

IMS database customization
IMS is customized differently from relational database management systems. The customization block for IMS looks different. Provide the IMS ID and the IMS SDFSRES library to access your IMS subsystem as a data source.

IMS access is broken down into two sections. The first section is used for CCTL or DBCTL support, which means you're simply making DLI transaction calls to IMS. The second section is for IMS Direct, which means we are accessing underlying files directly without making DLI calls. The IMS DBA can provide values to customize this block.

Adabas customization
Customizing Adabas as a data source involves enabling an initialization member by setting the MODIFY PARM NAME (ADABAS) to ‘YES’.

Configure the Started Task JCL

Configure the server, start a task JCL, and make more customizations. The high-level qualifier of either the target product installation libraries or the runtime libraries need be defined in the field HLQ. 

• Db2 for z/OS® requires the SDSNEZXIT and SDSNLOAD libraries to the STEPLIB
• IMS requires the IMS SDFSRESL library to the STEPLIB concatenation
• Adabas requires the ADABASE load library to the STEPLIB
• Ensure the SYSEXEC DD statement allocates the correct data set name from the AVZSIN00 member

Ensure that the SYSEXEC DD statement allocates the correct data set name that contains the initialization member from your customized sources and then attaches JCL.  The user-defined map data set needs to appear first and the concatenation in front of the supplied default map data set.

Configure the Command List (CLIST)

Configure the ISPF application to use the Member AVZ in the SAVZEXEC data set of the fully qualified load library from the runtime library you created in Step 5. You can also use the product target library created through the SMP installation process.

Invoke the ISPF panels by using screen commands.

• Edit hlq.SQVZEXEC member AVZ, and edit the following parameter for your product load library llib=’hlq.SAVZLOAD’
• To invoke the ISPF application, enter the following command on the ISPF command shell:
  EX ‘hlq.SAVZEXEC(AVZ)’ ‘SUB(AVZS)’

You can access the ISPF application to administer the server. Ensure the server is running before you invoke the ISPF application.

• Edit hlq.SQVZEXEC member AVZ, and edit the following parameter to that of your product load library llib=’hlq.SAVZLOAD’
• Copy the AVZ member to a data set that is allocated to the SYSPROC allocation for all TSO users
• To invoke the ISPF application, enter the following command on the ISPF command shell:
  EX ‘hlz.SAVZEXEC(AVZ)’ ‘SUB(AVZS)’

Verify the installation

Use the Eclipse-based DVM Studio Client to verify the installation by creating a Virtual Table to access your underlying data files as a relational database format using SQL.

The detailed checklist can be used to assist you in gathering the information you need to successfully install and configure the DVM Server.