Sizing queue processors and Stream data sets individually by changing the number of partitions of a topic
Each queue processor in Pega Platform is associated with a Stream data set in the Stream service. You can size individual queue processors and Stream data sets by changing the number of partitions of a topic associated with a particular queue processor or Stream data set.
For example, if you expect a potential increase in traffic for some queue processors during a product launch or the holiday season, you can increase the number of partitions for these queue processors.
- In Dev Studio, search for the pxAlterStreamPartitions activity.
- Click .
- In the Run Activity: AlterStreamPartitions dialog box, in
the stream field, enter one of the following
values:
- If the topic is associated with a queue processor, enter the name of the queue processor in uppercase letters.
- If the topic is associated with a Stream data set, enter the name of the Stream data set.
- In the partitions field, enter the desired number of partitions.
- Depending on the number of partitions, perform one of the following
actions:
- If you are decreasing the number of partitions, select the allowDataLoss checkbox.
- If you are increasing the number of partitions, leave the allowDataLoss checkbox cleared.
When you decrease the number of partitions, the system deletes the existing topic, and then recreates the topic with the new partition count. Existing data is lost. You need to select the allowDataLoss checkbox to confirm that data loss is allowed. When you increase the number of partitions, existing data is not impacted.
- In the upper-right corner of the Run Activity: AlterStreamPartitions dialog box, click Run.
Previous topic The support for default partition number customization Next topic Best practices for Stream service configuration