Below are listed general guidelines to optimize Campaign workflow performances, improve your workflow design and select the correct settings.
Adobe recommends you to create your workflows in a dedicated folder.
If the workflow affects the whole platform (cleansing processes for example), you can consider adding a sub-folder in the built-in Technical Workflows folder.
Because it makes them easier to find and troubleshoot if they are not performing in the expected ways, Adobe recommends to give your workflows proper names and labels: fill in the workflow’s description field to summarize the process to be performed so that the operator can easily understand it.
If the workflow is part of a process involving multiple workflows, you can be explicit when entering a label; using numbers is a great way to order the workflows (by Label).
You can configure the severity of a workflow in the workflow properties, in the Execution tab:
Providing this information when creating a workflow will help you understand the severity of the process configured.
This option has no functional impact on workflows other than campaign workflows.
Campaign workflows (workflows created as part of a campaign/operation) with a higher severity are executed in priority in case the campaign has many processes supposed to run simultaneously. By default, only 10 processes can run simultaneously in a campaign, according to the option NmsOperation_LimitConcurrency. For example, if a campaign contains 25 workflows, workflows with a higher severity will then be executed in the first pool of 10 processes.
All your scheduled workflows running on production environments should be monitored in order to be alerted if there is an error.
In the workflow properties, select a Supervisor group, either the default Workflow supervisors or a custom group. Make sure that at least one operator belongs to this group, with an email set up.
Before you start building a workflow, remember to define workflow supervisors. They will be notified by email in case of errors. For more on this, refer to Managing errors.
Regularly check the Monitoring tab to view the overall status of the active workflows. For more on this, refer to Instance supervision.
The Workflow HeatMap enables the Adobe Campaign platform administrators to monitor the load on the instance and plan workflows accordingly. For more on this, refer to Workflow monitoring.
You can copy and paste activities within a same workflow. However, we do not recommend to copy paste activities across different workflows. Some settings attached to activities like Deliveries and Scheduler could lead to conflicts and errors while executing the destination workflow. Instead, we recommended you to Duplicate workflows. For more information, see Duplicating workflows.
While developing your workflow, all activities will have a name, as will all Adobe Campaign objects. While the name is generated by the tool, we recommend you rename it with an explicit name when configuring it. The risk with doing it later is that it may interrupt the workflow with activities using the name of another previous activity. So it would be a difficult job to update the names afterward.
The activity name can be found in the Advanced tab. Don’t leave them named query, query1, query11, but give them explicit names such as querySubscribedRecipients. This name will show up in the journal, and if applicable in the SQL logs, and this will help to debug the workflow when configuring it.
Always start your workflow with a Start activity or a Scheduler activity. When relevant, you can also use an External signal activity.
When building your workflow, only use one Scheduler activity per branch. If the same branch of a workflow has several schedulers (linked to each other), the number of tasks to be executed will be multiplied exponentially, which would considerably overload the database. This rule also applies to all activities with a Scheduling & History tab. Learn more on Scheduling.
Use End activities for every workflow. This lets Adobe Campaign free up temporary space used for calculations within workflows. For more on this, refer to: Start and end.
To make spotting the workflow easier, we recommend using double dashes at the start and end of the activity label as follows: – My label --.
Most of the time, you will not know where the signal is called from. In order to avoid this problem, use the Comment field within the Advanced tab of the signal activity to document the expected origin of a signal for this activity.
A production workflow should not be updated directly. Unless the process consists of creating a campaign with template workflows, processes should first be tested on a development environment. After this validation, the workflow can be deployed and started on production.
Perform all testing in development or staging environments, not in production environments. Performance cannot be ensured in such a case.
Archived workflows may be kept on development or test platforms, in an Archived folder, but production environment should stay as clean as possible. Old workflows should be removed from production environment if they are inactive.
The Keep the result of interim populations between two executions option keeps temporary tables between two executions of a workflow.
It is available in the workflow properties’ General tab, and can be used for development and test purpose to monitor data and check results. You can use this option in development environments, but never use it on production environments. Keeping temporary tables could result in the size of the database increasing significantly and eventually the size limit being reached. Moreover, it will slow down the backup.
Only the working tables of the last execution of the workflow are kept. Working tables from previous executions are purged by the cleanup workflow, which runs on a daily basis.
This option must never be checked in a production workflow. This option is used to analyze the results and is designed only for testing purposes and hence must be used only on development or staging environments.
The Log SQL queries in the journal option is available in the Execution tab of workflow properties. This option logs all SQL queries from the different activities, and provides a way to see what is actually executed by the platform. However, this option should only be used temporarily during development and not activated on production.
Best practice is to purge the logs when they are not needed anymore. Workflow history is not purged automatically: all messages are kept by default. History can be purged via the File > Actions menu or by clicking the Actions button located in the toolbar above the list. Select Purge history.
To learn how to purge your logs, refer to this documentation.
Additional best practices should be applied on your workflows execution planning to avoid issues:
Improve your instance stability by implementing the following best practices:
Do not to schedule a workflow to run more than every 15 minutes because it may impede overall system performance and create blocks in the database.
Avoid leaving your workflows in a paused state. If you create a temporary workflow, make sure it will be able to finish correctly and not stay in a paused state. If it is paused, it would imply that you need to keep the temporary tables and thus increase the size of the database. Assign Workflow Supervisors under Workflow Properties to send an alert when a workflow fails or is paused by the system.
To avoid having workflows in a paused state:
Stop unused workflows. Workflows that keep running maintain connections to the database.
Only use unconditional stop in the rarest cases. Do not use this action on a regular basis. Not performing a clean closure on connections generated by workflows to the database impacts performance.
Do not perform multiple stop requests on the same workflow. Stopping a workflow is an asynchronous process: The request is registered, then the workflow server or servers cancel operations in progress. Stopping a workflow instance can therefore take time, especially if the workflow is running on multiple servers, each one of which must take control to cancel the tasks in progress. To avoid any issue, wait for the stop operation to be completed and avoid stopping a workflow multiple times.
In a production environment, avoid executing workflows in the engine. When the Execute in the engine option is checked in the Workflow properties, the workflow takes priority and all other workflows are stopped by the workflow engine until this one is finished.