Error in Parallel Processing

chrisf90 · April 12, 2016, 12:59

Hello everybody,

I am facing some problems with parallel processing. I run some simulations on a single machine mith multicores (I use 12). And everything is working fine until I switch on the DO radiation model. I have attached a screenshot of the Error message which occurs in TUI.

I am also wondering because i got other simulations running parallel on more cores previously also - also with DO model. It seems for me like a random problem, but I want it to work everytime.

Maybe it is a hardware issue but as I am not really a hardware specialist it is only a guess.

Advice would be highly appriciated.

Regards
Chris

LuckyTran · April 12, 2016, 16:01

Not sure what the problem or fix it. The error message just sates that the parallel cores couldn't communicate, which can be caused by anything. If this is recurring problem, then you would need to do some serious debugging and logging everything that you did to determine what the root cause is. If it's infrequent, just move along. Hopefully you get a more detailed error at some point to help you isolate the problem.

Maybe you can try reducing the number of divisions/bands in your model?

chrisf90 · April 13, 2016, 01:21

Thank you for your reply.

Alright, I will move on with some other cases in my parametric study and will see if the same or other errors will occur.
I will try that with reducing the number of divisions/bands in my models. But compared to other studies in literatur the number of those in my model are quite small (divisions 4x4; bands 2).
Maybe starting with small values and partly increasing could help?

However, I go to work right now and will see if I get it running. Thank you anyway at this point.

chrisf90 · April 13, 2016, 12:15

Quote:

Originally Posted by LuckyTran

Maybe you can try reducing the number of divisions/bands in your model?

Dear LuckyTran and other users

I tried to reduce the number of divisions and bands in my models and obtained that the problem occurs from the number of divisions.

Everything works fine until I increase theta and phi divisions over 1x1. Then the simulation crashes and states the error from post#1.

With this information - does anybody have hints for me which could help solving this problem?

Thanks in advance

April 13, 2016, 01:21		#3
chrisf90 New Member Christoph Ferk Join Date: Feb 2016 Posts: 17 Rep Power: 10	Thank you for your reply. Alright, I will move on with some other cases in my parametric study and will see if the same or other errors will occur. I will try that with reducing the number of divisions/bands in my models. But compared to other studies in literatur the number of those in my model are quite small (divisions 4x4; bands 2). Maybe starting with small values and partly increasing could help? However, I go to work right now and will see if I get it running. Thank you anyway at this point. Last edited by chrisf90; April 13, 2016 at 03:11.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Explicitly filtered LES	saeedi	Main CFD Forum	16	October 14, 2015 11:58
simpleFoam parallel	AndrewMortimer	OpenFOAM Running, Solving & CFD	12	August 7, 2015 18:45
problem for parallel processing	minaret	OpenFOAM Running, Solving & CFD	14	January 19, 2015 23:41
HP MPI warning...Distributed parallel processing	Peter	CFX	10	May 14, 2011 06:17
About parallel processing in Linux	tuks	CFX	10	August 8, 2005 08:22

April 12, 2016, 16:01		#2
LuckyTran Senior Member Lucky Join Date: Apr 2011 Location: Orlando, FL USA Posts: 5,674 Rep Power: 65	Not sure what the problem or fix it. The error message just sates that the parallel cores couldn't communicate, which can be caused by anything. If this is recurring problem, then you would need to do some serious debugging and logging everything that you did to determine what the root cause is. If it's infrequent, just move along. Hopefully you get a more detailed error at some point to help you isolate the problem. Maybe you can try reducing the number of divisions/bands in your model?