Time slot scheduling

A sweep can be scheduled to run within one or more time slots. A single time slot specifies the day of the week, the start time, and the duration that the sweep runs.

You can configure a sweep to run in one or more time slots. A sweep is effectively paused between time slots. For example, suppose that the expiration of the current time slot stops a sweep mid-way through processing a set of items. When the sweep runs in its next time slot, it resumes from where it previously stopped.

For run-once sweep jobs that process small database tables, time slots have marginal utility. But for processing large database tables, sweep jobs can benefit from time slots. Say that you want to update the retention date on a Document subclass that consists of several thousands of instances. You can configure a sweep job to run during non-peak hours, such as every day of the week between midnight and 2 AM. The job will run in only those time slots until it completes processing.

You can configure time slots for the sweep subsystem at the server hierarchy level (domain, site, virtual server, or server instance). This global configuration impacts all of the sweeps within that hierarchy. You can also configure time slots for individual sweeps, which override the time slot configuration in the sweep subsystem.

Time slots are not available for configuration when you create a sweep with a wizard. You must first create the sweep, and then open the sweep to configure it to run in time slots. Time slots are scheduled in the General tab of the sweep.

You cannot configure time slots directly on sweep policies. You configure the time slots on the policy-controlled sweeps to which the policies subscribe. Because a single policy-controlled sweep can be subscribed to by more than one policy, multiple sweep policies can run in the same time slots.