Set up and run in Control Room

Control Room manages your work data. Consider some of the following questions related to process automation:

  • How do I scale the workload efficiently over multiple execution environments? Control Room enables parallel processing. Need to run one step in the cloud, another on a Windows machine, and one more on a Linux server? Can do!

  • How do I rerun only the failed data items? I don't want to rerun everything. Control Room tracks the status of the work items. Successfully processed work items are tracked separately from the failed ones. This way, you can reprocess only the failed items if needed!

  • How can I be sure I don't lose any data if the robot crashes? By using work items, the status of your process stays up to date even if something horrible happens to your robot during execution.

  • How do I measure and communicate the value generated by the process? You can define the monetary value of your work items. This communicates the benefits of the automation in terms that business folks appreciate!

No need to invest in custom solutions. Instead of having to develop custom libraries, setting up databases, monitoring systems, analytics software, and figuring out how to use and maintain all that in the future, Control Room work data management provides a standard and best practice answer to all of these questions with a minimal effort from the user.

The local development has now reached a point where you feel that the robot is in good enough shape. Time to upload the robot to Control Room and set it up!

You create a new process in Control Room. You assign two steps. The first one is Produce traffic data work items. The second is Consume traffic data work items.

Running the process in Control Room

You start the process, and the robot gets to work - downloading the raw traffic data, transforming it and storing it as work items, and submitting the traffic data to the sales system API.

Some of the work items are completed successfully, some fail. You view the work items after the run has finished with an Unresolved state (meaning not all the work items were completed successfully).

┌──────────┐ ┌────────────┐ ┌──────────┐ ┌─────────────────┐ │ Producer │ → │ Work items │ → │ Consumer │ → │ Done work items │ → 💰 └──────────┘ └────────────┘ └──────────┘ └─────────────────┘ ↓ ↑ ┌───────────────────┐ ↑ │ Failed work items │ → → Retry → ↑ └───────────────────┘

Most of the failed work items have failed due to an application exception. Those you could retry, and they would eventually pass if the sales system API manages to process them.

One of the items has failed due to a business exception (in this case, due to the data validation failing).

You open the work item and see that the country code is longer than the expected three letters. You edit the country code by removing the extra letters and save the work item.

Then you select all of the failed work items and choose the retry option. Eventually, after a few retries, all the work items have been completed successfully. Yay!

In the future, Control Room will support auto-retrying the failed work items. Then you can focus on handling the business exceptions manually (if you want) and let Control Room and your robot take care of retrying the work items that failed due to application exceptions.