Compute node targeting. ALL broadcasts to every healthy node. Specific node IDs route to named nodes. Default: any single healthy node.
TARGET_NODES ALL
TARGET_NODES (node_1, node_2)The TARGET_NODES clause controls which compute nodes execute the pipeline run. When set to ALL, the scheduler dispatches the run to every healthy compute node in parallel. When a comma-separated list of quoted node identifiers is provided, execution is routed exclusively to those named nodes. When omitted, the scheduler assigns the run to any single healthy node using its default selection strategy. This clause interacts with MAX_CONCURRENT and PRIORITY to govern how runs are queued and distributed across the available compute fleet.