Control groups consist of groups of libraries, objects, folders, spooled files, and other types of data that share common characteristics. They also specify what kind of backup you want BRMS to perform, such as full, incremental, or cumulative-incremental. In addition, control groups specify the days on which the data will be processed, and the order in which the processing occurs. As such, control groups are the most important and useful element in the BRMS tailoring process.
Control groups can also provide flexibility in starting and stopping subsystems and holding job queues. Perhaps most importantly, control groups allow you to separate and save your most critical applications, which, in turn, allows you to more easily recover them. After you recover your most critical applications and data, you can recover the rest of your system. Your save strategy is likely to consist of multiple backup control groups. Sample Backup Scenarios: Working with Multiple Control Groups and Shared Devices provides specific information about how to design a backup strategy that uses multiple control groups.
The default control groups described in Backing Up Your Entire System do an excellent job of backing up your entire system. IBM recommends that you use these default backup groups, called *SYSTEM, or *SYSGRP and *BKUGRP, to perform your first backup and that, afterward, you leave them unchanged and use them on an as-needed basis. In addition, you can create some of your own control groups to better control related types of data.
Control groups often refer to, and work with, information contained in the function policies, for example, in the system and backup policies. Later in this section, you learn how to set control group attributes to override policy decisions.