Job assumptions

• The two jobs are a job for keeping track of hours worked and a job for settling business travel expenses.
• Two back-end servers are allocated to each job and are used exclusively for that job.
• During normal operations, the travel expenses job requires higher performance.
• At the end of the month, the workload of the work hours job increases.
• At the end of the month, the work hours job is given higher priority than the travel expenses job.

The following figure shows the back-end server configuration when load balancing is performed.

Figure 11-11 Back-end server configuration when load balancing is performed for Example 1 (two jobs, three units, four back-end servers)

Note

The number in parentheses indicates the switching priority of each back-end server within the unit. This is the same as the value of the standbypri operand of the standby server (standby is specified in the inital operand) for each server specified in the servers definition of the HA monitor for each server machine.

Note that (0) indicates the primary system. In this case, online must be specified in the inital operand of the servers definition for the applicable server machine or server and the standbypri operand must not be specified.

Explanation

• At the end of the month when the workload of the work hours job increases, back-end servers are migrated to balance the job load.
• This scenario is achieved by combining two basic scenarios for migrating back-end servers.
• When the priority shown in the figure is specified, you can prevent the load from becoming unbalanced among the back-end servers even if an error occurs after back-end server migration.

Job assumptions

• The two jobs are a job for keeping track of hours worked and a job for settling business travel expenses.
• Eight back-end servers are allocated to the work hours job and six back-end servers are allocated to the travel expenses job, and all back-end servers are used exclusively for their allocated job.
• During normal operations, the two jobs require the same level of performance.
• At the end of the month, the workload of the work hours job increases.
• Even at the end of the month, the performance of the travel expenses job must not be slowed down if at all possible.

The following figure shows the back-end server configuration when load balancing is performed.

Figure 11-12 Back-end server configuration when load balancing is performed for Example 2 (two jobs, four units, and 14 back-end servers)

Note

The number in parentheses indicates the switching priority of each back-end server within the unit. This is the same as the value of the standbypri operand of the standby server (standby is specified in the inital operand) for each server specified in the servers definition of the HA monitor for each server machine.

Note that (0) indicates the primary system. In this case, online must be specified in the inital operand of the servers definition for the applicable server machine or server and the standbypri operand must not be specified.

Explanation

• At the end of the month when the workload of the work hours job increases, back-end servers are migrated to balance the job load.
• This scenario is achieved by combining three basic scenarios for migrating back-end servers.