This topic describes the merge node concept, and how to create a merge node and define the merging logic. It also shows you the scheduling configuration and operation details of the merge node through a practical case.
- The merge node is a type of logical control family nodes in DataStudio.
- The merge node can merge the running states of upstream nodes, and aims to solve the issues of dependency mounting and running trigger of downstream nodes of branch nodes.
- The current logical definition of merge node does not support selecting nodes that are in the running state, but supports merging multiple downstream nodes of the branch nodes, so that more downstream nodes can be mounted to the merge node as a dependency.
For example, the branch node C defines two logically exclusive branches C1 and C2. Different branches use different logic to write to the same MaxCompute table. If the downstream node B depends on the output of this MaxCompute table, and must use the merge node J to merge branches first. Then add merge node J to the upstream dependency of B. If B is mounted directly under C1 and C2, at any given time one of the branch nodes will fail to run because it does not meet branch conditions. B cannot be triggered by the schedule to run.
Create a merge node
Define the merge logic
An example of the merge node
Run the task
When the branch meets the specified condition, select the downstream node of the branch to run. You can view the run details in the Running Log.
When the branch does not meet the condition and does not select the downstream node of the branch to run. You can view the node that is set to 'skip' in the Running Log.
The downstream node of the merge node is running normally.