Hi,
I executed parallel batch processing with the student_example and increased timeout up to 300 to see how the parallel batch processing will behave, because with default timeout of 30 a lot of scenarios were skipped by timeout before.
Now a lot of scenarios are skipped without any reason (see screenshot below). Is this behavior normal or did I something wrong in the configuration?
Regards Johannes
Hi Jode95,
Thanks for pointing this out, with student_example it makes sense to have many timeout as it can not find a solution within 30s, as for the scenarios processed without any reason in your screenshot, I think it is more like a multithreading problem, the result details will be corrected by the end. You can see this from astar, the first 6(num_worker_processes) scenarios have no result detail at the beginning and then are added in the end. Please let me know if there is still anything unclear.
Best,
Ling
Hi Ling,
yes I saw that the first 6(num_worker_processes) scenarios of astar have no result detail at the beginning.
In the screenshot I posted 10 scenarios were processed right now with 6(num_worker_processes) too, but only one scenario would been marked as “Solution not found”. The other scenarios that were processed already would not marked or counted for any status or reason, so I don’t know what’s the problem in this case…
Regards Johannes
Hi Johannes,
The reason might be those scenarios are still in the process loop and haven’t finished yet due to parallel processing, and also since you increased timeout limit, it took longer. I think actually those scenarios are processing and they are just not printed to the console immediately, as long as the final result summary is correct, this won’t be a problem. 