After creating a transform, I schedule it to run against my data files using policies. Policies are schedules that allow me to run transformations on an entire dataset in my cluster, and publish the repaired and enriched data to specified target locations. In Oracle Big Data Preparation cloud service, I navigate to the Policies page where all the existing policies are listed, and create a new policy. In the create policy page, I specify details to create the policy and details to schedule the policy. I enter a unique name for the policy and optionally, specify a description for the policy. I specify if I want to receive email notifications only when the scheduled policy starts and ends; or an elaborate email mentioning details of when each engine started and ended. Or, I can opt to not receive any emails notifications.
By default the policy is activated immediately after it is created. I select the transform that I want to run against my dataset in this policy, the directory where my source files are located, and the target location for the published results. The source and target directories could be Oracle Cloud Storage, Oracle Business Intelligence Cloud Service, or the default HDFS that is provided for this service. I specify how I want the files named for the published results each time the policy is run. In the Schedule section, I choose how often I want to run the policy, when the policy needs to run, and the start and end dates if the policy will run over a period of time.
After I save the policy, it is listed in the Policies page. Using the context menu, I can edit policy details or delete it. When the policy is run, I get a notification and a new job id is created. From the jobs page, I monitor the status of the policy. Once the job is completed successfully, I navigate to the Catalog page and download the published results to my local file system. In the Home page, the interactive dashboard displays a summary of all my publishing activities.