37.1.1 Function overview

If you execute an SQL statement or command on a base table, invalid row data might be added to a system table. If invalid row data continues to increase in a system table, the following consequences occur:

• Free disk space for storing system tables becomes insufficient.

• The increased number of pages in the DB area for system tables, which are referenced during searches, degrades the performance of the search function.

Executing the adbreorgsystemdata command reorganizes the system table data, thus freeing up space occupied by invalid row data. This processing is called system table reorganization processing.

The following figure provides an overview of system table reorganization processing.

Figure 37‒1: Overview of system table reorganization processing

Explanation:

System table reorganization processing is executed in the following two phases:

1. Phase 1

   System table data, with the invalid row data removed, is stored in an unload file (unload processing). Next, the data in the unload file is re-stored in system tables as reorganized data (reload processing).

2. Phase 2

   The pre-reorganization data is deleted, thus freeing up the space occupied by the invalid row data in the system tables.

The following explains the files:

• Temporary work files

  These are the work files that are created temporarily during reorganization processing of system tables. The created temporary work files are deleted automatically when the reorganization processing is finished.

• Unload files

  These are the files that are created temporarily during the reorganization processing of system tables. System table data, with the invalid row data removed, is stored in these files. The created unload files are deleted automatically when the reorganization processing is finished.

Note

For details about when invalid row data might increase in system tables, see Reason for reorganizing a system table in the HADB Setup and Operation Guide.

To Page Top