A HCL Link flow is a transactional data flow comprised of processing nodes.

A flow might be run on a scheduled basis, invoked by a REST API, ir it can run in response to some external event. Such an event might be the creation of a file, the arrival of a message on a queue, or the change of state of an object in an external application.

A flow passes data between processing nodes which perform operations on the data, and then output results to subsequent nodes in the flow. Nodes can perform these operations:

  • access external resources to get or send data
  • invoke HCL Link maps or flows
  • call REST APIs
  • format and convert data
  • conditionally route data

Extrememly high throughput speeds can be achieved by using the Split and Join nodes. These nodes allow data to be processed in parallel by creating batches and processing each batch in parallel.

Flows support sophisticated error handling whereby failures in the flow execution can result in earlier nodes rolling back their actions, such as leaving source data in place, rather than deleting it.