Error In Solver 32 bits and 64 bits versions

[naserhassania]

I have been running a model that I have developed for a shear wall structure on my laptop successfully and I was able to get the results that I need
When i tried running the same model on a computer at the lab it didn't work out. I use 32 bits version 2.5.0 on my laptop while the one in the lab is 64 bits.
Would that make any differences in the codes?
The error I got is due to the solvers, I have tried to change the system command and still got the same errors
The error I get is |:

OpenSees > source shearwall_frame_structure_DynamicAlgorithm.tcl
check 1
check 2
check 3
check 4
nodes Defined
check 5
check 6
wall elements are defined
Beam elements defined
column element defined
uniaxial material for beam hinges in defined
Beams Plastic hinges are defined
check 8
Column Plastic hinges are defined
check 7
Model is defined !!!!
WARNING BandGenLinLapackSolver::solve() -LAPACK routine returned 147
WARNING NewtonRaphson::solveCurrentStep() -the LinearSysOfEqn failed in solve()
StaticAnalysis::analyze() - the Algorithm failed at iteration: 0 with domain at
load factor 0.1
OpenSees > analyze failed, returned: -3 error flag
Model is defined and gravity analysis is completed !!!!
WARNING BandGenLinLapackSolver::solve() -LAPACK routine returned 147
WARNING BandGenLinLapackSolver::solve() -LAPACK routine returned 143
WARNING BandGenLinLapackSolver::solve() -LAPACK routine returned 151
ArpackSolver::Error with _saupd info = -9999
Could not build an Arnoldi factorization.IPARAM(5) the size of the current Arnol
di factorization: is 3factorization. The user is advised to check thatenough wor
kspace and array storage has been allocated.
WARNING StaticAnalysis::eigen() - EigenSOE failed in solve()

Even the analysis under gravity loads didn't run
Please any recommendations or suggestions on how to solve this error
I can send the scripts to check if needed
Thanks in advance for any help

[fmk]

32 or 64 bit should not matter .. could possibly be a round-off error, the 32 and 64 bit compilers i have are different versions of the intel compilers .. sound like a modelling issue if it is failing in first step gravity .. use a ProfileSPD solver and see if that works for gravity (i assume the system is positive definite at the start) .. some of the other fancy solvers swap rows and stuff in the factorization and even for models that have a rigid body can get a successfull solve sometimes, the ProfileSPD does not and so it catches bad models better.

[naserhassania]

Thank you sir for your quick reply, I will check this up and get back to you with the results

[naserhassania]

Prof. Frank

Well I have changed the solver option as you have suggested and I tried different ones too, and I still got the same error. I don't think it is failure due to the model definition, because I I have run the same model again many times on my laptop and it is working very well, just taking some time to run.
I need to use the computer at the lab because it is taking much time on my laptop and I need to do multiple times soon so the use of a computer with larger power is needed.
If I can please send you the model files and you have a look at them to figure out whats the problem.
what email I can send the model files to?

Thank you for your help and time
Regards
Naser

[fmk]

fmckenna ATTTTTTT berkeley DDOOTT edu

[naserhassania]

Sir, I have just emailed you with a zip file including the scripts for the model.

Thank you for your time and help

Regrads
Naser