*ERROR in u_calloc: error allocating memory

21

@rsmith
What do you mean by that?

Do you have any changes to my process regarding the use of *restart,read? Because I always felt like my operation wasn’t official. (delete the.rin file, change the resulting.rout file to a.rin file as the start file for the next.inp,)

22

@jbr
So how do I modify my program?

I’ve been feeling a little weird about my use of *restart,read (delete the.rin file, change the resulting.rout file to a.rin file as the start file for the next.inp), but I don’t know any other proper uses for *restart.

23

I would do the following:

  1. Start by creating a new mesh with only quad elements (S8 as @rsmith suggested).
  2. Make sure that nodes and elements are sequential and that no gaps between node/element numbers exist.
  3. Do a quick modal extraction, free-free or fixed, to make sure the model behaves as expected.
  4. Run your simulation as you have done in the past…
24

@jbr
Then I think my use of inp and *restart should be correct, because after 180+ runs, the memory allocation problem appears later…

25

did it have problems in latest versions also?

26

I would say that may be true but then something else is creating the issue, it is not a CalculiX problem…

27

I don’t know. I haven’t tried the new version yet,0.0

28

Then I’ll try to find out what might have caused it

29

Not sure if this could help but what about using OP=NEW to clean previous *CLOAD before the

*INCLUDE, INPUT=force.txt

"Applying a point load at a node in a direction for which a point load was

specified in a previous step replaces this point load, otherwise it is added. The

parameter OP=NEW on the *CLOAD card removes all previous point loads."

Not really sure about its effect on allocation for the geometric BC as it’s a dynamic analysis .

30

I’ve tried, but it hasn’t worked

31

I have tried the new version, but the problem still occurs, and I found that the problem is not caused by too little memory. I only got so far with the 2T of memory on my supercalculator that I got this error.

32

I found that the problem was not caused by too little memory. I only got so far with the 2T of memory on my supercalculator that I got this error. So could this problem be caused by some parameters set too small in calculix source code?