After you configure scheduling dependencies for an auto-triggered node, you can preview the dependencies to check upstream and downstream node relationships and view the estimated execution time for instances generated on a scheduling day. This helps you verify that your configuration is correct and ensure that nodes run as expected. If the preview does not match your expectations, adjust the scheduling dependency configuration to prevent execution delays.
Usage notes
-
If upstream and downstream nodes have different scheduling cycles, the dependency relationships can vary between cycles. To prevent unexpected behavior, preview instance dependencies before deployment, especially when the scheduling cycles or scheduling times of upstream and downstream nodes do not align. For more information about DataWorks dependency principles and examples for complex scenarios, see Must-read: Principles and examples for scheduling configurations in complex dependency scenarios.
-
The dependency preview is for reference only and may differ from the actual dependencies in the production environment.
NoteIf you modify the scheduling cycle of a node in the production environment, select Immediately After Deployment, and then deploy the node, the instances for the current day may be replaced. The dependency preview cannot reflect the actual instance dependencies for that day. For more information, see Instance Generation Mode.
-
The preview shows only the direct upstream and downstream dependencies of the current node.
-
For an accurate dependency preview, ensure all upstream nodes are saved.
-
In the dependency preview, solid lines represent same-cycle dependencies, and dashed lines represent cross-cycle dependencies (a dependency on the previous cycle).
Dependency preview diagram
DataWorks displays dependencies in three views: Not Aggregate, Aggregate By Workspace, and Aggregate By Owner. For more information about aggregation methods, see Appendix: Introduction to the DAG feature.
In the dependency preview toolbar, click the grid icon to switch to the Not Aggregate mode, which shows the complete node dependency relationships. Click the gear icon to switch to the Aggregate By Workspace mode, and click the person icon to switch to the Aggregate By Owner mode. When aggregated, upstream and downstream nodes are grouped into a single node that represents the corresponding workspace or owner.
Preview node dependencies
DataWorks generates a DAG of upstream and downstream nodes based on your dependency configuration. Use this graph to verify your scheduling dependency configuration.
Click any node to view its details.
On the Node Dependency tab, the node details include fields such as node ID, name, owner, and description.
Preview instance dependencies
If an auto-triggered node generates multiple instances on a scheduling day, DataWorks generates the upstream and downstream dependencies and planned execution times for all instances based on the node's dependency and time configurations. You can then verify that each instance's dependencies and execution time are as expected.
-
Click any instance to view its details.
-
For hourly or minutely nodes configured with time-based scheduling parameters, the parameter values are determined by the scheduling time of each instance. For example, if an hourly node uses the scheduling parameter $[hh24], the parameter for each instance is replaced with the instance's specific scheduling time. For more information, see Supported formats for scheduling parameters.
In the Preview Dependencies dialog box, click the Node Instance Preview tab. From the instance list drop-down list on the right, select the instance you want to view. For example, a node that runs every three hours generates eight instances (Instance 1 to Instance 8), corresponding to execution times from 00:00 to 21:00. You can then view the upstream and downstream dependencies and the planned execution time for each instance.