Execution Mode & Node Map Constraint

[jerome_rajan]

Is there any difference between the way setting 'Execution Mode' to sequential works and defining a job to run on a single node using a Node Map constraint?

I read that when the Execution Mode is set to sequential, the conductor does the execution . If true, how does the conductor do this?

I was thinking that the conductor is the master process that sets up the score, creates SLs and does all the logging activity.

[jerome_rajan]

Can someone please help me understand this difference or direct me to a sitee that can help me understand this?

Thank you.

[ray.wurlod]

Sequential execution is controlled by the configuration file. When the score is composed, the node used is a resource allocation exercise; usually sequential operators are allocated to the free-est available node from the default node pool (the node pool whose name is ""). Using a node map constraint gives you the flexibility to choose which node that is.

If you leave your conductor node out of the default node pool in the configuration file, sequential operators not constrained to execute in the pool in which the conductor executes will not be executed in that node, as they will execute on a node in the default node pool.