What is the maximum number of parallel loops in a single VI?

[bondhan]

Hi all,

I tried to find in the forum but I could not find similar questions. I am using PXI with 2 cores microprocessor. If I am using 2 parallel timed loops and each is assigned to each processor, then what happened if  add one more parallel loop, to which processor should I assign the loop? Or should I just leave it automatic? Is it better to have two loops only with all uncritical task in one of the loops (and make it bigger) or is it better to add one more parallel loop. What is the maximum number of parallel loops in a single VI?

Thank you

[Ben]

I suspect the max number of parallel loops is larger than anything a human is capable of coding. The old DSC Panel Wizard (now obsolete) used to drop loops for every object (a nightmare to support!).

But driving at the root of your question...

All TC operations should be in one processor and the non-TC in the other.

Make sure your loops "play nice" (i.e. do not dominate the core, do the min possible to stay TC and then go back to sleep).

THe Execution Trace Toolkit is helpful it determining who is not "playing by the rules" and why.

If anyone else would like to comment on this subject, I'd love to hear what Y'all have to say!

Just trying to help,

Ben