The default Engine Thread property of a data flow task is set to 5, is this the best setting? what if I would like to run complex data flow tasks on multi-processor machines, should I increase the engine thread? If so, then what is the recommended Engine Thread number for running complex data flow tasks in a multi processor system?
Even if i am running simple data flow tasks on a multi processor machine, should I change the engine thread?
btw, would you happen to know the maximum number of threads that can run in parallel? I'm going to put that in a guideline document that's why I need the numbers, after the actual package has been built, I can test and measure on the multi processor box we have.