You can use the Job Scheduler in the Orchestration Server to automatically start deployed jobs on your grid by using either time or event triggers.
You can think of the functionality provided by the time triggers as being similar to a distributed cron system (in fact, time triggers can be described in cron syntax). This triggering, coupled with the job control functions in the Orchestration Server, allows for the sophisticated automation of routine data center tasks.
For example, suppose you want to periodically harvest a large log file in a coordinated way from a farm of several hundred machines. First, you could create an Orchestration Server job that uses the datagrid for file movement. The job control options specify that the job should run on not more than three machines at once and sweep across the entire grid. You would then create a schedule to run this job at the desired interval.
As another example, you could use the Job Scheduler to trigger a discovery job every time a new resource is added to the grid. In this case, the job developer writes the discovery job to discover and set facts about the resource. Next, you would create a schedule to run this job on the RESOURCE_ONLINE built-in trigger. In fact, this type of triggered job is currently used in the standard set of deployed discovery jobs to detect specific resource CPU and OS information.
Yet another example would be to run a job on server startup that sends a notification e-mail to an administrator.
This section includes the following information: