Operator Guide

10.4.0
- 10.5
- 10.4.1

Back Next

Subscription Delivery Properties

The

Delivery

page of the Subscription wizard can include the following properties:

Data to Consume: Delivery scope and format of the data to consume. Choose one of the following options:
All available publications. Processes each published data set with a separate subscription mapping.
All available publications - aggregated. Groups and processes all published data sets and delivers a single data set. When you use an automatic mapping, the subscription sorts the data according to the publication date and time of the publication instances.
Latest publication only. Delivers only the latest published data set.

This section is disabled for unbound subscriptions.
Update Data in Target: Handling of data that exists in the target application. Choose one of the following options:
Append the new data to the existing data in the target. If the target type is Relational Database,
Data Integration Hub
adds rows to the table. If the target type is Flat File,
Data Integration Hub
appends records to the file.
Insert new rows and update remaining rows.
Data Integration Hub
inserts new rows and then updates rows that exist in the target.
Insert new rows and update changed rows.
Data Integration Hub
inserts new rows and then updates changed rows that exist in the target. Updating only changed rows slows down system performance. Applicable for subscriptions that consume data from topic tables where delta detection is enabled.
Delete rows that don't exist in the topic from the target.
Applicable if the type of topic to which the subscription subscribes is Full or if the subscription is an unbound subscription, and you choose to append new data or to insert new rows.

Overwrite existing data in the target. If the target type is Relational Database,
Data Integration Hub
truncates the target tables before it inserts new rows. If the target type is Flat File,
Data Integration Hub
overwrites the file.
Advanced Performance Tuning: Use the options in this section to fine tune the performance of data retrieval from the
Data Integration Hub
publication repository for subscriptions that use an automatic mapping. Performance tuning is important when you use unbound subscriptions.
Use pushdown optimization. Run querying and filtering on the
Data Integration Hub
publication repository instead of using a mapping with a PowerCenter Filter transformation.

Pushdown optimization offloads processing jobs from the PowerCenter repository to the publication repository. Because the publication repository is critical to system health, use pushdown optimization with caution.
In subscribers that subscribe to a topic that is managed on a Hadoop publication repository, basic filters are pushed to the Hadoop publication repository. Advanced filter expressions are not pushed to Hadoop and are processed in PowerCenter.

For more information, see the chapter "Pushdown Optimization" in the
PowerCenter Advanced Workflow Guide
.

Use PowerCenter partitions with ... partitions. Enter the number of partitions in the text field. It is recommended that the number of partitions equals the number of CPUs on the computer that runs the PowerCenter Integration Service.

The number of partitions you enter determines how many PowerCenter processes handle the request in parallel. While parallel processing is faster, it consumes more resources from the PowerCenter Integration Service. For more information, see the chapter "Understanding Pipeline Partitioning" in the
PowerCenter Advanced Workflow Guide
.

Sort by publications order. Sorts the data that the subscription consumes according to the order of batches by which the data is published. By default,
Data Integration Hub
does not sorts the data.

Rename Saved Search

Table of Contents

Operator Guide

Operator Guide

Subscription Delivery Properties

Subscription Delivery Properties