Data pipeline versus standalone workflows

Workflows in APEX AIOps Incident Management are processed in one of two ways:

As part of an ordered list of workflows executing sequentially. Workflows processed this way are called data pipeline workflows.
As an independent, standalone operation. Workflows processed this way are called standalone workflows.

Note

All event workflows are data pipeline workflows by default.

Quick comparison

The following table provides a high-level overview of the differences between data pipeline and standalone workflows:

	Data pipeline workflows	Standalone workflows
Workflow Types	Events, alerts, and incidents	Alerts and incidents
Invocation Methods	Workflow trigger	Actions Menu Run Standalone Workflow action Workflow API calls
Trigger	The workflow trigger defines the conditions that must occur for the workflow to execute	The workflow trigger only contains an optional filter
Actions	Must have at least one action Can contain multiple actions	Must have at least one action Can contain multiple actions
Rank	Workflows are ranked	Workflows have no rank
Scope Filter	Can use the `changes` scope filter field	No access to `changes` scope filter field
Delay Action	Can use the Delay action	Cannot use the Delay action
Execution	Sequential	Asynchronous

Data pipeline workflows

Data pipeline workflows run according to the order that they are listed on the page. Whenever a data pipeline workflow is triggered, it runs until it is completed and then starts the next workflow in the sequence, continuing until the sequence is finished. Events, alerts, and incidents processed by one workflow are then passed to the next workflow in the list for further processing (unless workflow actions which prevent this from happening are present).

The Priority column of the workflow view indicates the order in which workflows in the pipeline will be triggered. To learn more about priority, see .To change the order of workflows, see Change data pipeline workflow order.

Workflows processed in the data pipeline consist of a trigger and at least one action. The trigger defines the conditions that must occur for the workflow to execute. The action determines the change the workflow performs on the data. A data pipeline workflow can contain multiple actions that act upon the items triggering the workflow. Actions process data in the order in which they appear in the workflow configuration.

For a complete list of event, alert, and incident actions, see the Workflow action reference.

Standalone workflows

Standalone workflows are specialized workflows that are called individually, without triggering subsequent workflows upon completion. They are available for alerts and incidents, and are explicitly invoked through the following methods:

API calls (see example)
The Actions Menu (see example)
The Run Standalone Workflow action (see example)

Unlike typical data pipeline workflows, standalone workflows do not have a rank. They also do not have a workflow trigger or access to the changes scope filter field, instead relying solely on a filter that applies when the workflow is called.

Alerts and incidents processed by a standalone workflow are not passed on to subsequent workflows. To call another standalone workflow from an existing standalone workflow, use the Run Standalone Workflow action.

All standalone workflows must have at least one action. The action determines the change the workflow performs on the data. Workflows can contain multiple actions that act upon the items triggering the workflow. Actions process data in the order in which they appear in the workflow configuration.

You cannot use the Delay action in a standalone workflow. All other alert and incident actions listed in the Workflow action reference can be used.

Standalone workflows are asynchronous and run in parallel. Do not rely on the output of a standalone workflow in other workflows.

Why use standalone workflows?

Key reasons to use standalone workflows include:

Workflow API calls: Standalone workflows are useful when you need to execute a series of actions that cannot be accomplished with a single API call. For example, if a specific condition is met for an incident, you may want to change its description, add a tag, or perform other changes. Since no single API call can execute all these actions simultaneously, you can create a standalone workflow to do these actions and run it with the Workflow API.
Reusability: Standalone workflows are modular, and you can reuse them across your data pipeline. If you have a process that needs to be triggered at various points in your pipeline under different conditions, you can create a standalone workflow to encapsulate that process. You can then call this standalone workflow multiple times throughout your data pipeline.
Asynchronous processing: Standalone workflows are useful in situations where you need to perform actions independently from the main data pipeline. For instance, if you need to send alerts to an external system, this process may take time and could delay the execution of subsequent workflows in your pipeline. By creating a standalone workflow for this externalization process, you can initiate it at a specific point in your data pipeline without causing delays. This allows the data pipeline to continue processing other tasks while the standalone workflow exports the alert data asynchronously.

Additional information

For more information on standalone workflows, see also:

In this section: