Can an Embedded PXI Controller Multithread in parallel to PXI modules?

[MrManSplain]

I am not sure if this is a stupid question or not. I have been asked to spec and cost an ATE system with multiple positions to test >1 DUT at a time. Whilst I have done similar in the past there has always been a single expensive bit of COTS kit like a VNA/Spec Analyser etc. involved which had to be connected to each DUT in turn, i.e. each DUT tested sequentially one after the other, so no parallel testing was possible.

These DUTs have some digital lines, some analog, serial etc. but nothing that specialised. So my question is, assuming I am using an Embedded PXI Controller can this genuinely multithread and talk to PXI modules in parallel or is the PXI backplane limited to addressing one PXI module at a time?

[santo_13]

An embedded controller is just another computer like the one you use. You can run several applications in parallel. The bottleneck is your instrument, whether it can talk to several applications in parallel.

[MrManSplain]

Thanks Santhosh. Sorry I wasn't clearer in my question. I am aware an embedded PXI controller is just a (expensive!) computer with an operating system and therefore can inherently multithread.

The question I was trying to ask was can the PXI backplane handle parallel calls to different modules? or can only one PXI module be addressed, commanded and polled for data at a time?

If the answer is 'one at a time' then something like multiple separate multifunction DAQ units may be better as they would each have their own USB interface and therefore the PC could I believe, talk to them in parallel using its inherent multithread capability.

[santo_13]

If your question was about the PXI bus, it can talk to several slots in parallel, unlike the GPIB bus.

[MrManSplain · Accepted solution]

Ok excellent, thanks. I was trying to ascertain whether there could be a performance improvement using a PXI solution for a multi-DUT i.e. multi-position, test system assuming it perform tests for each position in parallel, and it seems that is the case.