Name

A logical, unique name for the task.

Command for Incremental Load

A table can be loaded in Full Mode or Incremental Mode. Full Mode refers to data loaded for the first time or data that is truncated and then loaded. Incremental Mode refers to new or changed data being added to the existing data.

The DAC maintains a last refresh timestamp whenever a table is changed during the ETL process. (You can view this timestamp under Design > Tables > Database Connections or Setup > Database connections > Tables.) If a table has a timestamp, the command appearing in this column is executed. If a table does not have a timestamp, the command for a full load is executed. If the execution type is Informatica, the workflow name is used as the command.

Command for Full Load

If a table has no last refresh timestamp, this command is executed.

Folder Name

For execution type of Informatica only. The folder in which the workflow resides. Note: The name cannot contain spaces.

Source DB

Source database that is queried.

Target DB

Target database that is populated.

Task Phase

Task phase of the ETL process. The DAC server uses the task phase to prioritize tasks and to generate a summary of the time taken for each of the phases.

Execution Type

Tasks are executed based on their execution type. The following types are supported:

• Informatica. Task is invoked on an Informatica server using pmcmd.
• External Program. Task is an operable program on the operating system where the DAC server is running. This program can be a batch file, shell script, or any other program that can be run like a bulk loader.
• SQL File. Task is a SQL script in .xml or .sql format. For more information, see Using SQL Files as an Execution Type in the DAC.
• Stored Procedures. Task is a stored procedure that has been defined on the databases.

In addition, there are several internal execution types that you will not be able to select when creating new tasks. These tasks are categorized as either internal change capture tasks or internal data warehouse tasks; all of these tasks are color-coded in pink in the Task Definitions tab.

• IMG_BUILD - internal change capture. If you are using multiple Siebel transactional sources, you cannot change the behavior of the change capture process. This task requires change capture tables to be created on the other sources also. When adding additional Siebel sources, go to Setup > Database Connections and select the appropriate Siebel source, right-click and select Change Capture Tasks. This action generates change capture tasks. Use this same action to disable or delete change capture tasks.
• IMG_SYNC - internal change capture. If you are using multiple Siebel transactional sources, you can create this task for the additional tasks for doing similar change capture sync processes. You cannot change the behavior of the change capture sync process. This task requires change capture tables to be created on the other sources also. This task should be used with discretion for Siebel sources only.

Execution Type

• QUERY_INDEX - internal data warehouse. This task allows you to alter when the Query indices are created. The DAC server drops all indices before loading when the CreateQueyIndexesAtTheEnd setting is set to True. When this setting is set to False, all the indices, regardless of the index type, get created as part of the task that does the loading.
• UPDATE_ETL_PARAM - internal data warehouse. This task is used for only to update W_PARAM_G from the DAC server. This task populates the preferences to the W_PARAM_G table in the Data Warehouse by querying values defined in the DAC repository. Because only one Data Warehouse per DAC repository is supported, this execution type should not be chosen for any task.

Build Image

Applicable for Siebel transactional sources only. When this check box is selected, the change capture for the primary/auxiliary source tables executes.

Truncate Always

When this check box is selected, the target tables are truncated regardless of whether a full or incremental load is occurring. Any indices registered for this table are dropped before the command is executed and are recreated after the command completes successfully. When indices are dropped and created, the table is analyzed so that the index statistics are up-to-date.

Truncate for Full Load

When this check box is selected, the target tables are truncated only when a full load is occurring. Any indices registered for this table are dropped before the command is executed and are recreated after the command completes successfully. When indices are dropped and created, the table is analyzed so that the index statistics are up-to-date. When the Truncate Always option is selected, this option is unnecessary.

Optional

Indicates that a task is optional and will be executed only when the table listed in the Conditional Tables subtab is included in an execution plan.

For example, the UpdateRecencyCat in Person Dimension task is optional and is associated with the Order Item fact table, which is listed in the Conditional Tables subtab. The UpdateRecencyCat in Person Dimension task will only be executed when the Order Item fact table is included in an execution plan.

Continue on Error

When this check box is selected, if the command fails, the dependent tasks are not stopped. However, if any auto-generated tasks fail, the dependent tasks are stopped.

Inactive

When this check box is selected, the record is inactive. Inactive tasks do not participate in the ETL process.

Edit

Allows you to edit the table selected in the top pane.

Description

Displays a description of the task selected in the top pane.

Source Tables

Displays the list of tables used for getting data for the task selected in the top pane.

The list contains the following column headings:

• Name. Name of the table.
• Primary. Indicates whether the table is a primary source of data.
• Auxiliary. Indicates whether the table is a secondary source of data.

Note: If a table is marked as Primary or Auxiliary and the Build Image property of the task is selected, the change capture process is invoked. There are special tasks that force the base table data to be extracted when data in auxiliary tables change.

A table can be neither Primary nor Auxiliary but still be used for getting some attributes to populate a dimension or fact table. The changes in these kinds of source tables are not reflected in the dimension or fact table once the data is populated.

Target Tables

Displays the list of tables where data is loaded for the task selected in the top pane.

The list contains the following column headings:

• Name. Name of the table.
• Primary. Indicates whether the table is a primary target table. For documentation purposes only.

Conditional Tables

Displays a list of tables that, if included in an execution plan, cause the optional task selected in the top pane to be executed.

For example, the Order Item fact table is a conditional table associated with the optional task called UpdateRecencyCat in Person Dimension. The UpdateRecencyCat in Person Dimension task is executed only when the Order Item fact table is included in an execution plan.

The list contains the following column headings:

• Name. Name of the conditional table.
• Inactive. Indicates whether the conditional table is inactive.

Depends On Tasks

Displays a list of tasks that are prerequisite for the execution of the task selected in the top pane. If any of the prerequisite tasks fail, the task selected in the top pane will be marked as Stopped.

The list contains the following column headings:

• Name. Name of the prerequisite task.
• Task Type. Type of task (read only).
• Optional Flag. When this flag is selected, the prerequisite task optionally depends on the task selected in the top pane. For example, task 1 loads table A, and task 2 loads table B and optionally depends on task 1. If an execution plan requires both tables A and B to be loaded, task 2 will wait for task 1 to successfully complete. If an execution plan requires only table B to be loaded, task 2 will be executed immediately, and task 1 will not be executed.

Dependent Tasks

Displays a list of tasks that are dependent on the task selected in the top pane. These tasks wait for the successful completion of the current task before they are executed.

The list contains the following column headings:

• Name. Name of the dependent task.
• Task Type. Type of task (read only).
• Optional Flag. See the description for Depends On Tasks.