PowerCenter Workflow Types
Data Integration Hub uses PowerCenter batch workflows for publications and subscriptions with automatic mappings.
You can use PowerCenter workflows for publications and subscriptions with custom mappings, publication pre-processing, subscription post-processing, and monitoring rules that invoke PowerCenter workflows. The type of PowerCenter workflow that you can use depends on the type of process for which you want to use the workflow:
- PowerCenter Batch Workflows
- You can use batch workflows for all processes.
- PowerCenter Real-time Workflows
- You can use real-time workflows for custom publications.
Batch Workflows
Batch workflows run according to a trigger and not continuously. Data Integration Hub uses batch workflows to process publications and subscriptions, to run publication pre-processes and subscription post-processes, and to perform actions for monitoring rules that invoke workflows. Batch workflows use the PowerCenter Web Services Hub to communicate with PowerCenter.
The following operations can trigger a batch workflow:
- •Scheduled publication or subscription starts.
- •Operator manually runs the workflow.
- •DX_Start_Publication transformation sends a request to the Data Integration Hub server.
- •DX_Notification transformation sends a request to the Data Integration Hub server to start subscription workflows for a completed publication or a request to start a post-processing workflow for a subscription.
- •A publication process or a subscription process meets the conditions of a monitoring rule that is configured to invoke a workflow.
When an publication or a subscription operation triggers a batch workflow, Data Integration Hub creates an event with a processing status. If the trigger occurs for a scheduled subscription, the event for the subscription has a delayed status from the time the data is ready to consume until the scheduled subscription start time. When a monitoring rule invokes a workflow, Data Integration Hub does not create an event.
Real-time Workflows
Data Integration Hub uses real-time workflows to process publications. Real-time workflows process data from real-time sources on demand. Real-time sources can be web-service providers, Apache Kafka, or Java Message Service (JMS) queues.
Real-time workflows are not started by Data Integration Hub. Running and maintaining the workflows is the responsibility of the Data Integration Hub developer. You can run real-time workflows continuously or on demand.
When publishing through a real-time workflow, you group the published data into single publications at predefined time intervals. At the scheduled publication time, Data Integration Hub creates an event with a processing status.