Documentation tree

Intelligent Scheduling

The Cantara Scheduler revolutionises the way that scheduled reports are managed which delivers a simpler, more reliable and maintainable scheduling solution for Oracle JD Edwards.  This is achieved by defining instances of a report version or Cantara extensions as a Request, where the Environment, Queue and recovery parameters are configured. The Requests are combined into logical Groups where the relationship between the Requests is defined, such as the order that they should launch and whether the Requests are dependent upon one another’s successful launch or completion.

Groups are also classified into Streams, such as Manufacturing or Finance or Companies, so that larger organisations can manage staff responsibilities or filter the Group records. A Group may consist of as few as one Request.

Dependencies and preferred launch triggers are defined on Groups, not individual requests. This is a whole new paradigm, and dramatically simplifies management of your Scheduler tasks and empowers you to create intelligent schedules.  Dependencies between Groups can be configured on whether the related Group has launched, has completed (success or failure), or has completed successfully, and works in conjunction with the scheduled trigger date/time. There is no dependency on single or multi-threaded queues, as the Cantara Scheduler monitors the launched reports, regardless of where they are running. With the addition of Cantara Extensions you can also include other actions such as executing AIS service calls or triggering OS level scripts as part of the schedule.

As an example, you can create a Group for all of your Manufacturing Accounting related nightly reports into a single Group, which is set to launch each report Request sequentially on QBATCH. A single trigger will initiate the submission of all of the report Requests, and subsequent Request in the Group will only launch when the current report completes. They do not need to be launched in a single threaded queue.

Once created, this Group can be treated as a single entity, which can have its triggers and dependencies revised as required if circumstances change.  Each Group can also be used to manage the actual launch time for dependent groups.

For example, we may not wish to launch the GL Batch Post report until all of the Manufacturing Accounting and all of the Overnight Sales reports have completed.

In the example below, the GL Batch Post Group could be scheduled to trigger just after the Manufacturing Accounting and Sales report Groups are scheduled.  If, however, this GL Batch Post Group is defined to have dependencies requiring both of these Groups to complete before it genuinely triggers, then it will wait until both Groups have completed, and then begin submitting the GL Batch Post Group’s report Requests.

Similarly you could set a dependency on the Repost Inventory Commitments Group definition, to require both the Manufacturing Accounting and the GL Batch Post Groups to complete successfully. If there is a failure within either Group, then the commitments would not be reposted, and an error message would be produced.

Variables can be defined to empower you to set one global variable value and utilise it in multiple Request and Group definitions.  This could, for example be the email address for all failure messages, or the date range for integrity reports etc.  Variables can be used in any Processing Option, Report Interconnect or Data Selection parameter associated with the Scheduler Requests and Groups.

The Cantara Scheduler provides controlled recovery, by allowing you to define if a scheduled Group should launch if it was unable to trigger at its scheduled time. You also define if a new instance of a Group should launch if the previous instance is still running.

Each report Group also has parameters for the Minimum Duration, Target Duration and Maximum Duration that the Group should run for.  If it completes in less than the Minimum Duration, or longer than the Maximum Duration, then it is considered to be a failure and an alert message is generated.  The Target Duration is used for the Gantt Chart, in conjunction with the Group dependencies to give realistic estimations of when a Group will be running.

Flexible Scheduling

Scheduling the planned trigger of Groups can be by Simple triggers, Cron Scheduling triggers or Period based triggers (utilising the selected Company’s date patterns).

You have the ability to override processing options, report interconnects and data selection values when the Report definition is defined, or for a specific Request.  Variables can also be used for any processing options, report interconnecs or data selection parameter to simplify maintenance of report requirements that change frequently or where a variable is updated for several different reports.

Scheduling Groups can be set to Active or Inactive so that you can control changes or prevent a Group from launching its report Requests

 

On This Page

  • No labels