Manage your service groups

Whichever service group you are using, it's best to plan out the values and setting for each, as described in the previous section. New entries should take into account the relative values of the other service groups.

As you will see, the panel to define each service group is somewhat different; therefore we show both, using the example.

Manage production service groups

To create Production Service groups that are almost identical, use the COPY command and then make any needed changes.

If the policy does not have any Production Service Groups you are taken to the Production Service Group Definition panel; if there are already PS Groups, then the following panel is presented.

You can create a new PS Group by using the File menu and 1. Configure New Production Service Group.

The E-Edit line command allows you to edit an existing PS Group, except for its name. To change the name, you must delete the PS Group and recreate it with the new name.

Whether creating a new PS Group, or editing an existing one, use the following panel to enter its values.

Complete the following fields:

• Production Service Group Name - this is editable for a new PS Group; otherwise it displays the name of an existing Production Service Group.
• Description - use this field to describe the purpose of the PS Group. It appears on the Production Service Group Management panel, in the Policy Report, and on the Job Summary Report for any job that is assigned to this service group.
• Production Importance - this value gives the relative importance of one job to another; values are 1 through 5, with 1 indicating highest importance.
• Late Job Acceleration - Specify Y for the SLM to use information supplied by a supported Job Scheduler to determine if a job is late.  If the job is late, it is assigned the Service Class and mapped to the next higher Production Importance value.
• Target - this is the time, in minutes and seconds, that would be the expected time this job would be in the queue on a "good" day. For supported schedulers, TM will automatically calculate the appropriate target for any given job. This is calculated from information the scheduler provides. Target is calculated as Job Start Time minus the Average Elapsed time. This allows the job an opportunity to begin execution in advance of its actual required start time.
• Acceptable - this is the time, in minutes and seconds, that a job assigned to this PS Group would be considered "late" if it were in the queue this long. Once the job reaches the Acceptable threshold it is immediately advanced to the Critical threshold. For supported Schedulers, TM will automatically calculate the appropriate Acceptable time for any given job. This is calculated from information the scheduler provides. Acceptable is calculated as Job Start Time.
• Generate Job Summary - Specify Y-yes to provide the Job Summary Report at the end of the output of reach job (SYSMSGs). It identifies:

–The node and system on which the job was read and then executed.

–The Policy in effect.

–The Service Group name and description.

–The actual and effective queue time.

–All delays and elapsed times, both in the queue and in execution.

Manage general service groups

If the policy does not have any General Service Groups you are taken to the General Service Group Definition panel; if there are already GS Groups, then the following panel is presented.

Each GS Group has a 2-level name, <control center> <type>. The set of service groups that have the same name as their first level identifier, are referred to by their mutual control center.

The E-Edit command allows you to edit an existing GS Group, except for its name. To change its name, delete the GS Group and recreate it.

The D-Delete command, used for a control center, deletes all the service groups in its set.

To create additional Service Groups, use one of the following options.

• C-opy an existing Control Center and all its Types, supplying a new Control Center name - the Type names remain the same.
• C-opy an existing Type supplying a name for the new Type within the same Control Center.
• create a new Control Center, and be prompted to add its first Type. Start with the FILE drop-down menu, option 2. Create New Service Group.
• create a new Type within an existing Control Center, and be prompted to supply the existing Control Center. Start with the FILE drop-down menu, option 3. Create New Service Group Type.

Remember to update your JAL to assign jobs to these new Control Centers and/or Types. See Create-your-job-processing-rules for details.

Whether creating a new GS Group, or editing an existing one, use the following panel to enter its values.

If you used the worksheet to determine your service levels, refer to it to fill out this panel.

Complete the following fields:

• General Service Group Control Center and Type - displays the name of an existing General Service Group.
• Description - use this field to describe the purpose of the GS Group. It appears on the General Service Group Management panel, in the Policy Report, in UDF displays, and on the Job Summary Report for any job that is assigned to this service group.
• BI (Batch Importance) - this value gives the relative importance of one job to another; values are 1 through 5, with 1 indicating highest importance.
• Aging Limit - indicates how far the jobs in this service group can progress in the queue. The default is A- Acceptable and these jobs stop aging at the Acceptable threshold. To allow the job to progress further in the queue, specify one of the C, C1, C2, or C3. A field will appear to for you to enter the relative value.
• Target - this is the time, in minutes and seconds, that would be the expected time this job would be in the queue on a "good" day.
• Acceptable - this is the time, in minutes and seconds, that a job assigned to this GS Group would be considered "still okay" if it were in the queue this long.
• Critical - this is the time, in minutes and seconds, that a job assigned to this GS Group would need escalation, based on its importance, if it were in the queue this long. This value only appears in concert with the aging limit settings.
• Generate Job Summary - Specify Y-yes to provide the Job Summary Report at the end of the output of reach job (SYSMSGs). It identifies:
  • The node and system on which the job was read and then executed.
  • The Policy in effect.
  • The Service Group name and description.
  • The level of service delivered, compared with its service time goals.
  • The actual and effective queue time.
  • All delays and elapsed times, both in the queue and in execution.