Working with Multiple Labels

You can assign multiple labels to engines to group them by a combination of geographic regions, environments, and projects.

Let's look at an example using Data Collectors. Assigning multiple labels to Transformers works exactly the same way.

You assign one group of Data Collectors the West and Finance labels, assign a second group the West and Marketing labels, and assign a third group the East and Finance labels.

When you select a label for a job, Control Hub can start the job on any Data Collector that is assigned the label. When you select multiple labels for a job, Control Hub can start the job on any Data Collector with all listed labels.

For example, you register four Data Collectors. You keep the default all label for each Data Collector and then assign the following labels to each:

Data Collector Labels
DataCollector1 all, Production, West
DataCollector2 all, Production, West
DataCollector3 all, Development, West
DataCollector4 all, Production, East
You do not configure the number of pipeline instances for any job - so by default Control Hub starts each job on one Data Collector assigned all labels specified for the job. Let's look at which Data Collectors can process a job, based on the labels assigned to the job:
  • If you select the all label for a job, the job can start on any Data Collector with the default all label: DataCollector1, DataCollector2, DataCollector3, or DataCollector4.
  • If you select the West label for a job, the job can start on any Data Collector with the West label: DataCollector1, DataCollector2, or DataCollector3.
  • If you select the Production and West labels for a job, the job can start on any Data Collector with both labels: DataCollector1 or DataCollector2.
  • If you select the Development and East labels for a job, the job fails to start because Control Hub cannot find a Data Collector assigned both labels.