Error after changing mesh size

General Questions
1

hi,

attaching 2 files only difference is mesh size

Model with smaller mesh throws error

running fine: https://drive.google.com/file/d/1E-CGOl5Rp-UX4GFa83oyYiDqDGahglYO/view?usp=drive_link

not running: https://drive.google.com/file/d/1CT1iYgisfvPPdqmcb5b2wlMHdMx4Gl7M/view?usp=drive_link

new to PPM, please help

2

You have to set access to those files/folders on Google Drive to “Anyone with the link” before copying the links and pasting them here. Please replace the links in your post once you do that.

3

sorry,

updated links

4

It seems that both files have the same mesh consisting of 234543 elements and 421038 nodes. What is the error message that you get and when does it occur?

5

Decascading the MPC’s

Determining the structure of the matrix:
Using up to 3 cpu(s) for setting up the structure of the matrix.
number of equations
8025114
number of nonzero lower triangular matrix elements
320737998

increment 1 attempt 1
increment size= 1.000000e+00
sum of previous increments=0.000000e+00
actual step time=1.000000e+00
actual total time=1.000000e+00

iteration 1

Using up to 3 cpu(s) for the stress calculation.

Using up to 3 cpu(s) for the symmetric stiffness/mass contributions.

Not reusing csc.
±------------------------------------------------+

  • PaStiX : Parallel Sparse matriX package     +

±------------------------------------------------+
Version: 6.0.1
Schedulers:
sequential: Enabled
thread static: Started
thread dynamic: Disabled
PaRSEC: Disabled
StarPU: Disabled
Number of MPI processes: 1
Number of threads per process: 3
Number of GPUs: 0
MPI communication support: Disabled
Distribution level: 2D( 128)
Blocking size (min/max): 1024 / 2048

Matrix type: General
Arithmetic: Float
Format: CSC
N: 8025114
nnz: 649501110

Job failed - no results exist.

Process elapsed time: 399.566 s

6

For me, it keeps (slowly) going. I would try with Pardiso instead of PaStiX (solver can be changed in step settings, Pardiso is better for bigger meshes) and maybe using standalone CalculiX since there’s a chance there will be more specific errors.

P.S. The simulations in both files were completed successfully on my PC.

7

ok. thanks for quick support
i will try with Pardiso. i will keep you posted once i have the results.

8

This seems to be my default answer to any Pastix-doesn’t-work problem but it looks like you’re using mixed precision which is pretty bad because it sometimes causes wrong results. That could mess up nonlinear.

Arithmetic: Float