Sticky Subagents for Orders

Current Situation

• With JS-1954 Subagent Clusters are organized for one of following scheduling modes:
  • fixed-priority: always use the first Agent and in case of unavailability only switch to the next Agent
  • round-robin: execute each next job with the next Agent
• With JS-1986 Subagent Clusters support the following scheduling mode:
  • all-priority: an order is split into child orders that execute with all Subagents configured for a Subagent Cluster.

Desired Behavior

• New Block Instruction for workflows: a number of instructions such as jobs can be put inside a Block Instruction that specifies the preferred Subagent Cluster to be used. Jobs within a Block Instruction that make use of the same Subagent Cluster as the Block Instruction will stick to use of the same Subagent. 
  • Similar to a round-robin cluster an order will start to execute the first job with the next available Subagent.
  • Any subsequent job executions caused by the same order are executed on the same Subagent (fixed-priority).
• If the Subagent becomes unavailable then the order remains in a blocked state.
  • A blocked order continues once the Subagent becomes available again.
  • Users can suspend and resume the blocked order from its current node or from a previous node. The Subagent Cluster will then select the next available Subagent to process the order.