AX2249
You can use Scheduler to perform batch processing of files. There are two options to perform batch processing in Scheduler:
-
You can set up a report file for batch processing, using a Batch Control Sheet. Then in Scheduler, you use a File Processing task to run the single report that is set up for batch. That batch report then processes all the files in the list.
OR
- You can set up multiple File Processing tasks in a Scheduler job, each one processing a different report (or the same report with different settings). In this scenario, no Batch Control Sheet is used—effectively, the Scheduler job defines the batch.
In most cases, it is more intuitive and flexible to set up a file for batch processing, rather than use multiple File Processing tasks. Using the Batch Control Sheet, you can easily see all of the files in the batch and review their settings side-by-side. You can list a file collect report as the last file in the batch, to collect the results of the prior files in the batch. Additionally, you have the option of processing the batch manually, without Scheduler.
Generally, only certain edge cases will gain advantages from using multiple Scheduler tasks to define a batch. For example:
-
You can process the reports in parallel instead of consecutively. When using the Batch Control Sheet, reports are processed consecutively, in the order listed. If you instead create a batch by using multiple Scheduler tasks, you can configure the tasks to run as parallel sub-jobs, which may improve the performance of the overall job. This would only be a significant advantage if the batch contained several reports with lengthy processing times.
-
You can easily override any file processing setting for any file. When using the Batch Control Sheet, the only "built-in" override settings are for the multipass source column and the source filter. It is possible to override other settings, but you must set up batch variables to do so. If you instead use File Processing Scheduler tasks for each report, you can directly override any file processing setting.
Batch processing and file collect
If batch processing is used to generate snapshot copies of various reports, you can use file collect to combine those snapshot copies into report packages and then deliver the packages to the appropriate recipients (and/or save the packages to a file share).
There are a variety of ways that you can incorporate file collect as part of a batch process. The most common method is to create a file collect report (using the file collect option of file processing), and then list that report as the last file in the batch. This way, the entire batch process—from generating snapshots, to creating report packages, to delivering the packages—is contained within a single file that can be processed manually or via Scheduler.
If desired, you can use any of the alternate methods for incorporating file collect. For example:
-
If you are using the multiple-Scheduler-task method of creating a "batch," the last task in the job can perform the file collect operation. This task can be either a File Processing task that processes a file collect report, or it can be the Scheduler-specific Collect Worksheets task.
-
You could still use a batch report to define the files to process, but maintain file collect separately. When running the batch report via Scheduler, you could use either the Collect Worksheets task or a separate File Processing task to perform the file collect. For example, you might want to separate the file collect operation from the batch list if you wanted to process multiple batch reports and then have only one file collect process at the very end. If your file collect process is set up to dynamically generate file lists, then you could optionally include or exclude reports or entire batches, and still generate file collect packages.
For more information on the differences between using file collect reports and the Scheduler-specific Collect Worksheets task, see Using Scheduler to perform file collect.
