Heuristic Queue Allocations

The Heuristic Queue Allocations view lists all scheduled Jobs, for the date selected, that are assigned to a Generic Queue A Generic Queue represents one or more Execution Queues. Each Execution Queue points to a system where ActiveBatch jobs run. Each system must have an ActiveBatch Execution Agent installed on it. When a Job that is associated to a Generic Queue is triggered, it will run on one of the Execution Queues that is a member of the Generic Queue. One of the key purposes of a Generic Queue is it supports Execution Agent high availability (active-active). Should one Agent and its assigned Execution Queue go offline for maintenance purposes or for unknown reasons, the surviving Execution Queue members of the Generic Queue can pick up the load. When a Job is associated to a Generic Queue, it means can run on more than one system - that is, it can run on members of the Generic Queue. On a per Job basis, you can set Job properties limiting the Execution Queue members the Job can run on, if desired. that has the Heuristic Queue Allocation (HQA) property checked. When HQA is enabled, scheduled Jobs are pre-assigned to run on a system-selected Execution Queue An Execution Queue identifies the system where a Job will run. The system is specified in the Execution Queue's Machine property. The system entered in this property must have an ActiveBatch Execution Agent installed on it. When a Job is associated to an Execution Queue, it will always run on the system specified in Machine property, unless a manual override is performed (for example, using an instance's "requeue" operation) (that is a member of the Job's associated Generic Queue) based on historical analysis and machine learning.

 

Note: Heuristic Queue Allocation is a checkbox property located on a Generic Queue's Properties sheet.

 

HQA allows scheduled workflows to be assigned from a Generic Queue to a set of Execution Machines that optimizes the use of the underlying machines and the relationship between the machine and the elapsed time of the specific Job within the workflow.

 

 

The view consists of five (5) columns: Name (Path), Expected Begin Execution, Expected Duration, Queue (Execution Queue assigned) and the Job template's Generic Queue assignment.

 

In the above image, the Jobs are assigned to a Generic Queue named GenQueue, as per the Template Generic Queue column. The Queue column is the system-selected Execution Queue each Job is set to run on.

 

 

The above figure for the HQA View shows a selected Job and the four (4) operations that are available (after right-clicking on the Job):

 

• Change Queue - This operation allows you to change the pre-allocated Queue to one of your choice.

• Reset Queue - This operation allows you to reset the Queue allocation back to what HQA had specified.

• Find Template in Tree - This operation finds and selects the Job definition in the Object Navigation tree.

• Find Queue in Tree - This operation finds and selects the Queue definition in the Object Navigation tree.

 

If the “Change Queue” operation is selected, a new dialog will be displayed with all the Execution Queue members of the Generic Queue, allowing you to select one to override the HQA computed one.