Set up Parallelization for Derivation Tasks (Obsolete)

Step type: Simple user interface.

Step description: This step splits the derivation path execution into partitions, which enables you to parallelize the execution.

WARNING This function is obsolete. As of version 23.09, the partitioning of derivation tasks is set by the step Generate tasks for parallel derivation builder.

Detailed functions: The following parameters are available:

  • Path object: Select option to restrict the parallelization on a specific derivation path.

  • Update generated tasks: If preselection tasks have already been generated, this option removes any existing partitions and builds them again. No new tasks are generated.

  • Partitioning options: Allows you to use the key with the best granularity and select the number of partitions. Note that to achieve an acceptable level of performance, the selected fields must be part of an existing database index when running on AnyDB (as opposed to SAP HANA).

    • Partition by all key fields: Chooses the table key to generate DTS partitions.

    • Partition by selected fields: Provides a list of custom fields that is used to generate DTS partitions.

    • Remove existing partitions: Removes the partitioning from the derivation task assigned to the path object above.

    • Field name: Editable only together with the option Partition by selected fields.

    • Number of partitions: Specifies the number of expected partitions. This parameter first counts the rows of the table to determine the partition size.

    • Rows per partitions: Specifies the number of rows per partition. This parameter overrides the parameter Number of partitions and determines the partitions according to the entered number of rows. If you leave this parameter empty, the parameter Number of partitions is used.

  • Display button: Shows the WHERE conditions of the created partitions.