The Database › Elements of the Database › Structure of the Checkpoint Data Set

Structure of the Checkpoint Data Set

The checkpoint data set is divided into two sections of equal size.

Note: Be sure to allocate an even number of cylinders for the checkpoint data set):

• The first section holds actual checkpoint data.

  The first four blocks of the checkpoint data section contain changed block masks for up to 32 different operating systems. These change masks enable each operating system to read only those blocks that have changed since the last time it accessed the checkpoint and to write only those blocks that it changes.

• The second section holds recovery data. This data is used if your system crashes while it is writing the changed blocks.

  Data in the recovery section is recovered automatically.

Each section of the checkpoint data set can hold a maximum of 3,872 blocks.