Data Integration Tasks in Data Integration Hub
You use Data Integration tasks in Data Integration Hub publications and subscriptions with custom mappings.
In publication tasks, the publishing application is the source of the task and the Data Integration Hub publication repository is the target of the task. In subscription tasks, the Data Integration Hub publication repository is the source of the task and the subscribing application is the target of the task.
When you develop a Data Integration task for a Data Integration Hub publication, you select the Data Integration Hub connection as the target object. When you develop a Data Integration task for a Data Integration Hub subscription, you select the Data Integration Hub connection as the source object.
Note: You do not use Data Integration tasks in data-driven publications and subscriptions.
Data Integration Tasks Rules and Guidelines
When you develop Data Integration mappings and tasks to use in Data Integration Hub publications and subscriptions with custom mappings, consider these rules and guidelines.
General rules and guidelines
Consider the following rules and guidelines when you create Data Integration mappings and tasks:
- •Do not run tasks that you create for Data Integration Hub from within Informatica Intelligent Cloud Services. You must run the tasks from Data Integration Hub by running the publication or the subscription to which the task is associated.
- •When you use the Data Integration Hub connection, the target object in a publication mapping or task and the source object in a subscription mapping or task presents the list of topics defined in Data Integration Hub. The format of the list is TopicName.tableName.
- •Data Integration Hub determines the scheduling of the publication or the subscription based on the settings that the operator defined for the publication or the subscription. When you create the Data Integration task, in the Schedule page of the task wizard, verify that the option Do not run this task on a schedule is selected.
- •To distinguish between publication tasks and subscription tasks, indicate the type of the task in the task name. When you select a task for a publication or for a subscription, you can easily select an appropriate task.
For example, name a publication task Pub_<TaskName>, and name a subscription task Sub_<TaskName>.
Synchronization task rules and guidelines
Consider the following rules and guidelines when you create synchronization tasks and mappings:
- •The task operation for publication tasks is an insert operation.
- •When you create a publication task, select the Data Integration Hub connection as the target. When you create a subscription task, select the Data Integration Hub connection as the source.
- •Synchronization tasks do not support multiple sources. Therefore, when you create a synchronization task for a publication or a subscription with multiple sources, create a relationship between the sources for the following use cases:
- - Publications: when you publish from multiple tables.
- - Subscriptions: when you subscribe to multiple tables, or when the subscription is a compound subscription.
Mapping task rules and guidelines
Consider the following rules and guidelines when you create mapping tasks and mappings:
- •The mapping operation is an insert operation for both publication and subscription mappings.
- •When you create a publication mapping, select the Data Integration Hub connection when you configure the target properties. When you create a subscription mapping, select the Data Integration Hub connection when you configure the source properties.
For instructions on creating Informatica Intelligent Cloud Services mappings and tasks for Data Integration Hub, see the Data Integration Hub Connector Guide.
Using Intelligent Structure Model in Data Integration Hub
Data Integration Hub supports unstructured flat files to publish data. You can use intelligent structure models to structure data from unstructured or semi-structured files and create custom cloud mappings that use intelligent structure models and publish unstructured data to Data Integration Hub.
For more information about how to use intelligent structure model in Data Integration Hub to publish unstructured or semi-structured data refer to the H2L ...