PartitionClustering

Attributes

This parameter has the following attributes:

Usage

Use this parameter to enable or disable the distribution of work between multiple agents based on the physical partitions of a target table (as defined in the DBMS). Select one of the following values:

• Specify Yes to process all of the partitions in a target table with a single agent.
• Specify No to process each partition in a target table with a separate agent. The default value is No.

For more information, see Distribution-of-work. Be aware of the following points when you specify this parameter:

• High-speed Apply Engine ignores this parameter when it distributes work between multiple agents based on units of recovery (when the DistributionType parameter is UR).
• BMC Software recommends setting this parameter to Yes if any SQL statement in your input updates the partitioning key so that a row in the target table moves from one partition to another. If you set the parameter to No in this situation, you increase the possibility that High-speed Apply Engine can execute SQL statements in an incorrect order or that multiple agents can encounter deadlock conditions.
• If you specify No, exercise caution when the partitioning key of a target table is not the same as the primary key. High-speed Apply Engine can encounter synchronization errors between the agents that process different partitions. To avoid synchronization errors, define logical partitions (with [ObjectPartition] parameters) based on the table’s primary key.
• High-speed Apply Engine supports the standard range partitions available on Db2 for mainframe targets.
• High-speed Apply Engine supports the standard range and hash partitions available on Db2 LUW targets on UNIX or Windows.
• High-speed Apply Engine supports range partitions available on Oracle on UNIX or Windows. High-speed Apply Engine does not support hash, list, composite range-hash, or composite range-list partitioning.