Issue with some processors converging and others not?

[epratt]

I have code that, in part, consists of a "while" loop that applies a known displacement at one of 14 nodes on a column. The sp $node 2 command is specific to only the nodes defined for each processor.

The script runs great, and matches my OpenSees single processor version exactly, until at one step the solution doesn't converge. I have a subroutine where if the analysis doesn't converge, the step increment is reduced, and the program starts back at the top of the While loop. I think the problem is that some of the processors converge, and others don't. I have tried to use "barrier" commands to force all the processors to sync at strategic points, but then the program just hangs for infinity waiting for the "converged" processors to catch up.

Have struggle with this for days now. Again, my single processor version works great, just slow. I would love to share the .tcl files, but I can't seem to find where I am able to attach files on this forum?

Please help?! Don't have much hair left at this point...

Erin

[epratt]

Thanks Noxic for this totally inappropriate response.

Can anybody help with this issue? Again, will be glad to send the tcl files, just can't attach them to this forum.

Erin

[mhscott]

Spammers are all over this message board.

You can book a 1-on-1 session for OpenSees debugging help on consulting projects: https://www.silviasbrainery.com/one-on-one-training