Skip to content

Running and Managing Jobs

Use the Run and Manage Jobs page to select a model and execute the job. On the top half of the page, you can select a model, environment configuration, and input and output folders. You can also specify the amount of time you want the job to run before terminating it to save unnecessary processing costs.

The bottom half of the page shows the jobs you have run with their status, model name, environment configuration, and date and time created. You can also view the details by clicking the Details button.

Run and Manage Jobs Page Illustration

The following image shows the Run and Manage Jobs page.

Run and Manage Jobs Page

How to Run a Job

Follow these steps to run a model job:

  1. Click Run and Manage Jobs on the Model Execution menu. The Run and Manage Jobs page opens.
  2. Select an entry from the Model field. Required.
  3. Select an entry from the Environment Configuration field. Required.
  4. Enter the number of hours you want the job to run before terminating in the Max Run Hours field. This prevents jobs that never completed successfully from using unnecessary processing time.
  5. Click the Select button in the Input Folder field. The Select Input Folder dialog box opens.
    Input folder
  6. Choose an entry for the Input Folder field from the list, click the Actions button for that row, and click Select.
  7. You can also open a folder to view its contents by clicking on the Actions button and selecting Open.
  8. Click the Select button in the Output Folder field. The Select Output Folder dialog box opens.
  9. Choose an entry for the Output Folder field from the list, or open a folder to view its contents by clicking on the Actions button and selecting Open.
  10. Click Run. If the job executes successfully, a Job Created message displays.

How Job Manager Uses Your Input Parameters

In Predictive Learning, all jobs require one of these parameters:

  • Data Exchange
  • Internet of Things (IoT)
  • Integrated Data Lake (IDL)
  • Predictive Learning Storage (PrL Storage).

For the first three parameters above, Job Manager stores the copied input in a temporary location and ensures the location is available to your code, when the code executes.

Job Manager will copy the entire folders (recursively) and data files from Data Exchange, IoT and Data Lake. For the IoT exports the actual content will be a parquet file, which consists of a subset of multiple files. When you write scripts that you plan to be executed in a job and you require the input parameters, always use the variables provided by the job execution contexts: inputFolder, outputFolder and datasetName. You can find more details about how to use these parameters in the Using Inputs from Job Execution section.

Job Details Illustration

The bottom half of the Run and Manage Jobs page lists all the jobs you have run, and allows you to open and view job details. The table lists the status of each ob, the model name, environment configuration used, and the date and time the job was created.

You can view more information on the Job Details page, pictured here.

Job Details Illustration

How to View Jobs and Details on the Manage Jobs Page

Follow these steps to view the details of a job on the Run and Manage page.

  1. Navigate to the Run and Manage Jobs page. The page opens.
  2. Click the Details button in the Actions column for the job you want to view. The Job Details dialog box opens.
  3. View the information and click Cancel to close.

Last update: January 22, 2024