Workspace Node: Big Data Analytics Model Execution - Specifications - Run Model Tab

Cluster name

The list box displays the available clusters in the server selected on the Specifications - Select model tab.

1. 1. Click the Update list of available clusters button to update the list.
2. 1. Start a job on server (Synchronous). If nothing has changed with the model, ( if the model still has the same sink information on both the node and server), select this option button to start a new job on the server. The node maintains a synchronous connection, and waits for results until it completes.
 If the run fails, an error message is displayed and no downstream results are generated.
 If the model has changed on the server, i.e., If the model sink(s) have changed, an error message about the change is displayed, and you are prompted to update the model list before running it.
  
3. 1. Connect to a running job on server (Synchronous).
 If there is no current job to connect to, and the If no running job exists, start a new job on server (Synchronous) check box is selected, the node starts a new synchronous job with the server. If the If no running job exists, start a new job on server (Synchronous) check box is not selected, the node reports that there are no running jobs to connect to.
4. 1. Start a job on server (Asynchronous).
      • If there are no jobs started on the server (the node maintains a job ID and checks for status of an already started job if it exists).
      • If nothing has changed with the model, (the model still has same sink information on both the node and server), a new asynchronous job is started on the server.
      • If the model has changed on the server ( If the model sink(s) have changed), an error message about the change is displayed, and you are prompted to update the model list before running it.
      • If there is already a job started, and it is still running on the server, the status of the job is reported. If the already started job has completed its run, this option starts a new asynchronous job.

Use existing results

Define an age limit in days, hours, minutes, and seconds for the selected sink (result) file in the node.

A result file is valid if the difference between the current date and the last modification date of the file in HDFS does not exceed the age limit set.

• If the results are valid, a downstream node is generated.
• If the results are invalid (i.e., older than specified age, do not exist on server or are stale.

You can specify whether to Show a warning dialog and prompt user to start a job on server (Asynchronous), Automatically start a job on server (Asynchronous), or Report errors and stop processing.

For synchronous job runs, Poll server for job status every _ Seconds. Set how often to display job status, in seconds.

When starting a job on server warn user about erasing existing sinks (output files).

Select this check box to display a warning when existing output files will be overwritten.

Run node as a task on Statistica server

• When this check box is selected, the options that throw warning dialog boxes will be disabled.
• If this check box is cleared, the scheduled task will wait on a warning dialog box that needs to be addressed in order to proceed further.

Options

The Options button menu is available regardless of which tab is selected.