Interrupting Nested For Loops/Sequences

[jaysmall]

I have a lagre program consisting of nested for loops and case statements calibrating an array of patch antennas. The entire process of which will take about 15 hours. During this time, I want to have two things happen.

Every 20 minuites or so, I want the process to stop what it's doing, follow certain commands for the measurement equipment to recalibrate itself, and then go back to what it was doing.

I also want to protect it against unpredicted and predicted crashes. If the system needs to stop for whatever reason, I would like to be able to pick up the process where it left off. A sort of checkpoint system, where, when it passes a checkpoint, if it needs to be restarted, I can tell it to start from the last checkpoint it passed.

Anyone had any experiece with these types of problems?

[MikeS81]

Hi jaysmall,

i mean no harm, but it sounds like you can useTest Stand for it. You will have the ability to write a sequence, to start parallel processes and you can define points, where to start after stoping/breaking the sequence. See also this example of a test sequencer.

Hope it helps.

Mike 

[jaysmall]

Unfortunately I don't believe we have a license to TestStand. Is there another way I could do this?

[Ray.R]

You would need to appropriately architecture the code.  Are you familiar with State Machine representation?  Basically describe how the software needs to behave based on events.  From that you can develop the appropriate architecture in Labview and code your solution.

R

[smercurio_fc]

You can read this for more:

Application Design Patterns: State Machines

Message Edited by smercurio_fc on 08-29-2008 09:31 AM

[paul_a_cardinale]

Try this.

[smercurio_fc]

That VI was written in 7.1 and uses a private method which is not supported. There is no documentation on its operation and if it doesn't work you can't complain about it. It should not recommended to be used for production code.

[ShotSimon]

---

jaysmall wrote:

I also want to protect it against unpredicted and predicted crashes. If the system needs to stop for whatever reason, I would like to be able to pick up the process where it left off. A sort of checkpoint system, where, when it passes a checkpoint, if it needs to be restarted, I can tell it to start from the last checkpoint it passed.

---

Could you open and append parameters to a file? Every time you restart execution of the vi after crashes you can look through the contents of the file to determine the last operation that was successful and continue with the last know good settings.  -SS

[jaysmall]

Keeping track in a file seems like a doable idea to keep track of where I am in the sequence. Make a state machine whose states are kept track of in a file like a checkpoint tracker.