Overview of extract

Option 2, Extract, enables you to extract data from one or more related objects (Db2 tables or views and/or MVS files), including Db2 referential integrity-related data and user-defined application-related data that are loaded in the specified File-AID/RDX relationship file. Should your extract request include relations to an Output Key file or external IMS related data, File-AID/RDX generates an Output Key file. For IMS Output Key files, File-AID/RDX also allows a BMC AMI DevX File-AID for IMS Extract to be executed as a step following the File-AID/RDX extract using the IMS key file. Or you can use the resulting key file separately with BMC AMI DevX File-AID for IMS to extract the IMS related data.

File-AID/RDX enables you to control the scope and depth of your extract by specifying extract criteria, including which relationships to exclude from the extract, how many rows to extract, and what selection criteria should be applied to the Db2 tables and/or MVS files to be extracted. You may define multiple selection criteria sets to be saved in the extract request file. Thus, you can create complex criteria and store the related extracted records in a single extract file. Furthermore, you can specify an Input Key file as additional selection criteria for the Db2 or MVS driving object during the extract process.

The Extract function consists of an interactive definition phase and an execution phase.

The interactive phase takes place online. In preparation of the execution phase, it allows you to specify information about the following:

  • Driving object
  • Relationship file
  • Navigation, sampling, and selection criteria
  • Extract file, Extract Request file
  • Input Key and Output Key files
  • Data Privacy options
  • Unrelated objects
  • Checkpoint/Resumption
  • Single Db2 table extract with Selective Columns and/or Delimited Extract file.

Your current profile provides default values for all new extract requests. It also controls the starting panel for the interactive phase and its flow. For more detail on how to select and set up your profiles, see Profiles and Global Defaults.

Of course, you can also create new extract requests from previously saved ones or simply repeat old requests without any changes.

In addition, File-AID/RDX builds JCL that you can edit and save before you submit it for execution. You can also create a single request member that combines the extract and the load specifications. For more information about how to create a combined batch job, see Continue With Load and Load.

Normally, the execution phase takes place in batch mode. Your installation may also permit interactive (online) execution of the extract, however, you should consider it for short extracts only. During the execution phase, File-AID/RDX can copy the extracted data to any of the following destinations:

  • A new or existing sequential file;
  • A new or existing member in a new or existing PDS

Important

The maximum data record length (LRECL) that can be extracted into the extract file is 32,748 (max. of 32,756 minus 4 bytes RDX data, minus 4 bytes VB data).

File-AID/RDX will issue a message when an object file has a longer LRECL, but allow you to start the extract. However, when a record to be extracted exceeds 32,748, the extract will fail.

Extract processing has been enhanced to utilize Db2 for z/OS Multi-row fetch feature providing potential reduction in CPU time, Service Units and improved performance. When extracting from tables with LOB or XML column types, multi-row fetch processing will not be used unless these columns are excluded from the extract.

During extract processing, File-AID/RDX uses temporary files. For information about temporary file naming conventions and allocation values, Temporary File Defaults.

File-AID/RDX produces reports that summarize the extract results at the end of the execution phase.

This section provides information about the following topics:

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*