What have we introduced?

We have implemented infrastructure changes to discover IBM Db2 Analytics Accelerator Loader Server for z/OS from IBM Db2 Administration Foundation for z/OS. You can use the Db2 Administration Foundation user interface changes to load object data to multiple accelerator servers with Db2 Analytics Accelerator Loader Server. Here, the data is loaded from a virtual table or virtual view.

• The 'data server' is the Accelerator Loader Server that hosts the virtual table or view.
• The 'target server' is the Accelerator Loader that has access to the target Db2 subsystem and the accelerator.

This blog will guide you on how to install, configure, and use this new feature in the Db2 Administration Foundation.

Let’s get going!

What to install?

You need to install the following:

• Db2 Administration Foundation for z/OS and apply APARs PH54968 and PH54452.
• Db2 Analytics Accelerator Loader for z/OS version 2.1 and apply APAR PH54984.
• Db2 Analytics Accelerator Loader Server for z/OS version 2.1 and apply APAR PH60437. This APAR includes the following:
  • Member HLVBIN4 in the hlq.SHLVBIN library
  • JCL HLVPLCPY in the hlq.SHLVCNTL library
  • Two configuration files HLVPCUST and HLVPPROD in the hlq.SHLVSAMP library
    

1. 1. Run the JCL by performing the following: 
      1. Replace '&&HLQ.SHLVBIN' with the dataset name where the binary file HLVBIN4 is present.
      2. Replace '&&your/directory/path' with the directory (plugin_hlv_top_folder) that will be used to store the binary files.
      3. Submit the JCL HLVPLCPY. This creates the directory, sets its permissions, and extracts the binaries from HLVBIN4.
         
   2. Configure Db2 Analytics Accelerator Loader Server by performing the following: 

• • • Modify the YAML file HLQ.SHLVSAMP (HLVPCUST) by providing values in the following fields: 
      

PathToProductYaml: DSN:<Yaml file dataset>
IZPConfigurationDirectory: <plugin_hlv_top_folder>/izpconf
IZPPluginJarDirectories: - <plugin_hlv_top_folder>/izpjars

These file paths in the HLVPCUST are required for the IBM Db2 Administration Foundation to discover the Db2 Analytics Accelerator Loader Server.

How to configure?

You need to configure UMS Zowe for Db2 Analytics Accelerator Loader Server to load data from Accelerator Loader Server objects to multiple Db2 Analytics Accelerator Servers in parallel. Perform the following configuration steps:

      1. Navigate to the ZWEYAML member of PARMLIB.

      2. Enable discovery by setting components.izp.toolsDiscovery.enabled to true. 

      3. Specify the locations of the Db2 Analytics Accelerator Loader YAML files in components.izp.toolsDiscovery.discoverySearchPaths: 

- "DSN:hlq.SHLVSAMP(HLVPCUST)"

For example,

toolsDiscovery:                          
  enabled: true                          
  discoverySearchPaths:                  
  - "DSN:HLQ.SHLVSAMP(HLVPCUST)"

      4. Restart Zowe. 

           Zowe is configured for Db2 Analytics Accelerator Loader Server.

      5. Verify whether Tools Discovery has recognized and loaded the YAML file for Db2 Analytics Accelerator Loader Server by validating the message that appears in the Zowe log.

           For example,

<discoverToolsIfEnabled> : Yamls processed during discovery: {5639-OLE={2.1.0=<ToolsDiscoveryEntry pathToProductYaml=<hlq.SHLVSAMP>(HLVPCUST), productYaml={product_id=5639-OLE, product_name=Db2 Analytics Accelerator Loader Server, product_ver=2, product_rel=1, product_mod=0, RVT=184}, instances=>} : Tools Discovery completed.

Notes: 

• The user ID under which the Zowe server is running should have READ access to the data set hlq.SHLVSAMP(HLVPCUST). To set up resource access permission, see Setting up users and teams.
• Zowe must be restarted to apply the changes made to hlq.SHLVSAMP(HLVPCUST) data set.

How to use the feature in Zowe?

      1. On the navigation menu, click Explore > Objects.

      2. Select Accel Loader server from the list of subsystem types on the top right of the page. 

      3. Specify the search criteria, such as the object name, object type, and subsystem, and click Apply.

Note: If you don't specify any object types or subsystem, all object types or all subsystems will be searched.

      4. Select  of the required tables displayed in the result table and then click Load to accelerators.

      5. Select the required values and check boxes for the following fields:

Section	Field	Description
Target information	Target Db2 subsystem	Select a Db2 subsystem from the drop-down to list the accelerators available for the selected subsystem. The name you choose must match with the registered Db2 subsystem.
	Table creator	Specify or select a table creator from the drop-down.
	Table name	Specify or select a table name from the drop-down.
	Available accelerators	Select the check box of one or more accelerators from the list of available accelerators displayed. If an accelerator is associated with a group, the Group label is displayed next to the accelerator name Note: You can select up to eight accelerators to load data simultaneously.
Create table DDL options	Database name	Specify or select the database name to use when loading data to the accelerator. This is an optional field.
	Table space name	Specify or select the name of the table space to use when loading data to the accelerator. This is an optional field.
	CCSID	Use this option to override the default CCSID of the target database. This option generates a CCSID clause on the Create table DDL statement for the target table. Select one of the values from the drop-down list: ASCII, EBCDIC, UNICODE, or leave the option blank to omit the CCSID clause from the generated DDL. This is an optional field.
	Accelerator only table	Select the checkbox if this table is used to load data only to the accelerator and not to simultaneously load data to Db2 for z/OS.
	Enable unicode column expansion	Select the checkbox to increase the precision of maximum length columns by a factor of 3. This option ensures that the conversion of EBCDIC to UNICODE cannot cause an overflow condition due to character expansion. When the checkbox is clear (default), the DDL for the target table on the source character columns is VARCHAR(n), where n is the maximum size of the source EBCDIC column.  When the checkbox is selected, the DDL for the target table on the source character columns is VARCHAR(n*3), where n is the maximum size of the source EBCDIC column. The SELECT statement is adjusted accordingly in the generated JCL
	Enable loader parallelism	Select the checkbox and enter the number of parallel tasks for parallel processing. The default degree of parallelism is 4.
	DDL Preview	Click the drop-down to preview the SQL CREATE statement used to create the accelerated table.

      6. Click Next. 

      7. Select the required values and check boxes for the following fields:

Tab	Field	Description
JCL generation	Target server group	Specify the name of the target server group. This is an optional field.
	Utility ID	Specify the target server utility ID. This is an optional field.
	StepLib DD concatenation libraries	Specify the data set names of the Accelerator Loader load libraries that you want to use: Accelerator loader library (1) Accelerator loader library (2)
	Db2 load libraries	Click the drop-down to view the field values that are auto-populated. You cannot edit these values.
	Db2 dynamic sql program (DSNTEP2)	Specify the required details for running the DSNTEP2 sample Db2 program:  Db2 load library (RUNLIB) Plan name (the default is DSNTEP2) Note: These fields are optional.
Tab	Field	Description
JCL settings	Job name suffix	Specify a suffix to generate a default job name. This one-character suffix is appended to the user ID. For example, A.
	Job account	Specify the optional accounting information that you can add to the job statement. For example, you can either leave this field blank or enter a value.
	Execution class	Specify the execution class to be used in the job statement. For example, A.
	Message class	Specify the message class to be used in the job statement. For example, X.
	Region size	Specify the region size to be used in the job statement. For example, 0M.
	Temporary DASD name	Specify the generic unit name to be used in the job step for allocating temporary work files. For example, SYSDA.

      8. Click Next. 

      9. Select the required values for the following fields:

Section	Field	Description
Job creation	JCL library	Specify or select the name of the JCL library from the drop-down. This is a mandatory field.
	Member name	Enter the name of the member to store your generated JCL.
	Job name	Enter the job name.
	Execute job	Select the check box to execute the job that is generated and saved. If the check box is not selected, the job will be generated but not executed. Note: After the job is executed successfully, the job ID is displayed to check the status.
Job step details	Drop table	Select the checkbox to drop the existing target table definition. Note: Select this checkbox only if the accelerated table already exists.
	Create table	Select the checkbox to create a new target table definition.
	Load	Select one of the following options from the drop-down: LOAD REPLACE – Replaces the data in the table. This option adds the REPLACE control card to the generated JCL.  LOAD RESUME – Loads new data to an existing table. This option adds the RESUME YES control card to the generated JCL.  Note: Selecting LOAD RESUME disables the Drop table and Create table check boxes.
	Set current SQLID	Specify an alternate authorization ID under which the Drop table and Create table statements are executed in the generated Accelerator Loader job. This is an optional field.

9.       10. Click Submit. A message is displayed that the job has been submitted successfully along with the job name and job ID.

Conclusion

You can use the ‘Load to accelerators’ feature to load data from Accelerator Loader Server objects to multiple Db2 Analytics Accelerator Servers in parallel. This loads data from the virtual table or virtual view.