Job Management Partner 1/Automatic Job Management System 3 System Design (Configuration) Guide

[Contents][Glossary][Index][Back][Next]


6.1 Database reorganization

This section describes how to estimate when maintenance of the embedded database needs to be performed.

When JP1/AJS3 is used over an extended period with the embedded database serving as the scheduler database, invalid areas are created in the indexes as records are erased from the database. You must clean up these invalid areas by either of the following methods:

Using the ajsembdbreclaim command, you can reclaim and reuse pages in the database that consist only of invalid areas (unused pages). This command can be executed even while the server is running. By executing the command periodically, you can enable better use of data areas. In most circumstances, if you use this command you will not need to reorganize the database.

However, when page usage efficiency declines (for example, pages have only one item of valid data), the number of pages that can be reclaimed declines and it might not be possible to reuse them efficiently. The database must then be reorganized.

Organization of this section
(1) Flow of maintenance tasks
(2) Estimating when to perform maintenance

(1) Flow of maintenance tasks

The following figure shows the flow of maintenance tasks.

Figure 6-1 Flow of maintenance tasks

[Figure]

(2) Estimating when to perform maintenance

With an embedded database, you must either execute the ajsembdbreclaim command, allowing invalid areas to be reused, or reorganize the database at regular intervals. To estimate when to execute the ajsembdbreclaim command or reorganize the database, you need to analyze the database areas. Work out how often maintenance tasks will be required, based on the analysis results described below.

When estimating when to reorganize the database, consider on which day usage of the database peaks, and by how much usage increases, on a weekly and monthly basis.

The following describes how to analyze database areas.

(a) Analyzing database areas

Execute the ajsembdbstatus command to check the database status.

If the analysis result of the ajsembdbstatus command shows that there are few unused segments and many unused pages, reorganize the database.

For details about the ajsembdbstatus command, see ajsembdbstatus in 2. Commands in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 1.

(b) Messages indicating an area shortage

When the data storage efficiency of the embedded database declines, message KFPH00211-I or KFPH00212-I is output to the Windows event log or UNIX syslog. If either of these messages is output, analyze the database status by executing the ajsembdbstatus command, and reorganize the database as required. Examples of the output messages are shown below.

Examples of messages indicating an area shortage
KFPH00211-I RDAREA usage xxx%, RDAREA="xxxxx" xxxxx
KFPH00212-I Table should be reorganized, RDAREA="xxxxx", AUTHID=xxxxx, TABLE=xxxxx
 

If either of these messages is output immediately after you execute the ajsembdbreclaim command or reorganize the database, you will need to expand the indicated RD area. For details about how to expand an RD area, see ajsembdbaddarea in 2. Commands in the manual Job Management Partner 1/Automatic Job Management System 3 Command Reference 1.

[Contents][Back][Next]


[Trademarks]

Copyright (C) 2009, 2010, Hitachi, Ltd.
Copyright (C) 2009, 2010, Hitachi Solutions, Ltd.