hexagon logo
Register Sign in
  • State Not Answered
  • Replies 3 replies
  • Subscribers 4 subscribers
  • Views 65 views
  • Users 0 members are here
Options
Related

Error: Dynamic memory allocation failed. There is not enough memory available.

GPiove
I try to run a simulation but after a while it stops and I get this error:
"Cannot create dynamic subarray EUL EL FLOAT VARS.                      
       Dynamic memory allocation failed. There is not enough                  
       memory available. This can occur when your computer is                 
       heavily loaded with other applications."
I close all the application but the problem still exist, I try also with another pc but the error is the same.
I have a 32 GB ram and 900 GB free space in the disk. I can't find where is the problem.
Is it possible that I have to change some parameters?
Attached Files (2)
  • 0
    Hi,
    Sometimes the memory estimator struggles to correctly determine the amount of memory required to load the model during initialisation.
     
    To fix you need to add the MEMORY-SIZE line after the START right at the top of the dat file, or in Patran in the Execution Control Parameters in the Integer and Float Memory Size fields.
     
    To find the values you need, take a look in the OUT file after such a type of error (any Dynamic memory allocation failed error):
     
    The WORDS USED will be the value it is trying to use, while the WORDS DEFINED is what is actually used. In my example above I have defined the integer and float size both to 50000000 to be larger than what is needed.
     
    So in the dat file this would show up as:
    MEMORY-SIZE=50000000,50000000
     
    It usually takes a few iterations to find the values to initialise with. But once you've done this for your typical large models, then you typically reuse the same settings.
  • 0
    I can't find the WORDS USED in the .OUT file, anyway I try with different value but the problem remain
    Attached Files (1)
  • 0
    Hi, it seems it crashed so did not wrap up the final messages, but in the dat file I see you were trying to run with 640 million Euler elements (4000x400x400). Feasibly you should look at <0.5 million Euler elements for model debugging.
© 2025 Hexagon AB and/or its subsidiaries. Privacy Policy | Cloud Services Agreement