When we execute a job on the execution server, will that job still require resources from the Talend server itself. Does the Talend server only generate and deploy or does it also play a role in the execution of the job too? Should the execution server be where we concentrate our hardware resources?
Yes, the Execution Server(s) are where the jobs run. This is where you can scale out Talend, by having multiple execution servers ( say to isolate cpu/memory utilization to specific processes ) or scale up with a single larger server.
The generate takes place wherever you have the command line service(s) installed.
What are the limitations of scaling out on execution server can we still have job level dependencies across execution server? How is this distribution of resources managed in a scale out scenario?
