Ultra Tasks

Ultra Tasks offer the speed and scalability needed for critical integrations that demand high availability, high throughput, and continuous execution. This feature ensures that data reaches its destination instantly, regardless of data volume, endpoint variety, or integration complexity.

Ultra Tasks enable pipelines to continuously consume documents from external sources. This includes sources incompatible with Triggered Tasks or those requiring low-latency processing.

Key features:

  • Continuous processing: Ultra Tasks allow pipelines to run continuously, ideal for always-on pipeline executions.
  • Scalability and load balancing: An Ultra Task can manage and load-balance multiple Ultra Tasks instances across multiple Snaplex nodes, processing multiple documents simultaneously.
  • Reliability: The Ultra Task monitor ensures the necessary number of instances are running, automatically restarting tasks if an instance fails or a node needs a restart.
  • Efficient queue management: Ultra Tasks pull requests from the FeedMaster queue and assign them to the associated Ultra Tasks instances.
  • Support for listening Snaps: Ultra Tasks can also be based on pipelines using a listening Snap, such as JMS Consumer, which runs indefinitely.

Prerequisites

  • Your Environment (Org) needs to have a subscription to the Ultra feature.
  • The Snaps in the associated pipeline for an Ultra Task need to work in Ultra-mode.

Support and limitations

  • The pipeline requirements for Ultra Tasks limit the use of certain Snaps. Refer to Snap Support for Ultra Tasks for details.
  • When you access the URL for an Ultra Task that is in an Environment different from the one you last selected, Classic Manager displays the Task details, but the Environment does not automatically change to the one that owns the Task. Use the selector at the top right in Classic Manager to switch to a different Environment.
  • Headless Ultra does not support the Auto-scaling setting for Instances per Snaplex because a FeedMaster node is not required.

Known issues

  • In some cases, Ultra Tasks duplicate messages that are being processed from the FeedMaster node, resulting in failed Retry errors. If you have this issue in your Environment, refer to Enabling the Prevention of Duplicates Enhancement.
  • After changes are made to a pipeline, the instances will restart. Ultra Tasks restart the pipelines so that multiple instances might run at the same time after making changes to the pipeline.