Skip to Content

Create Workflow (part 3), Build a Task Workflow in SAP Data Intelligence, Trial Edition

Build a task Workflow to combines both tasks which you created during the previous tutorials into a process which you execute at once by using SAP Data Intelligence, trial edition.
You will learn
  • How to build a task workflow to combine graph using the Pipeline operator
roy-zhongRoy ZhongJune 4, 2020
Created by
anishgupta85
October 19, 2018
Contributors
anishgupta85
roy-zhong

Please note here in this tutorial GCP refers to Google Cloud platform and AWS refers to Amazon Web Services and Azure refers to Microsoft Azure.

  • Step 1
    1. Log on as the system user in the default tenant

    2. In the Application Launchpad page open Modeler.

    3. Create a new graph.

    4. Add Workflow Trigger operator, 2 X Pipeline operators and a Workflow Terminator to the graph by drag and drop.

    5. Connect operators.

      • Connect the output out port of the Workflow Trigger to the input in port of the first Pipeline operator.
      • Connect the output out port of the first Pipeline operator to the input in port of the second Pipeline operator.
      • Connect the output out port of the second Pipeline operator to the stop in port of the Workflow Terminator operator.
      Workflow example
    6. Configure operator.

      • Right click on the first Pipeline operator and go to Open Configuration.
      • Under the parameter Graph name select the graph that you have created in the tutorial Create Workflow (part 1), Enrich data with Data Transform. In this case, you have named it as mytest.workflow1.
        u
        Pipeline configuration
    7. Similarly, for the second Pipeline operator, select the graph that you have created in the tutorial Create Workflow (part 2), Aggregate data with Data Transform under the parameter Graph Name. In this case, you have named it as mytest.workflow2. Also increase the parameter Retry interval from 20 to 200 for both the Pipelines.

    Log in to complete tutorial
  • Step 2
    1. Once done, save the graph. In this case, you have named it as as name mytest.workflow3 and description Workflow 3.

    2. Now execute the graph using the button at the top of the page.

    3. The execution status is displayed in the bottom part of the screen and it changes from Running to Completed once the execution completes.

      View result

    You have now executed the Workflow 3 but you will notice in the above screenshot that, Workflow 3 calls the linked pipeline, Workflow 1. Once the execution of Workflow 1 completes, it starts the execution of Workflow 2.

    Remark: When you look at the EnrichDevices or STATISTICS_DATA data set, you will not see any changes compared to the previous two tutorials.

    What to do if Workflow 2 and 3 are in status Dead

    • Workflow 1 generates a randomly generated filename for each execution. For that reason any subsequent attempt to run the graphs in this tutorial will fail because Workflow 2 has not been updated to use the new filename. To fix this issue refer to Part 2 of this tutorial -> Step 2 -> Point 10.
      Dead graphs

    Which operator would you use to assign already created workflow to run in current pipeline with other workflows?

    Log in to complete tutorial
Back to top