Abaqus

Brief Description:

Abaqus is a general purpose finite element analysis package. It can be used to simulate the response of structures and solid bodies to loads, impacts and thermal stress and to visualize the results of simulations.

Abaqus is available to all NCSA academic users on the following systems:

• Xeon/Linux (tun.ncsa.uiuc.edu), version 6.6-1
• SGI-Altix/Linux (co.ncsa.uiuc.edu), version 6.6-EF1.

Commercial use of Abaqus is governed by the NCSA Private Sector Program. Please contact Seid Koric at skoric@ncsa.uiuc.edu for more information. 

Interactive Access:

Interactive limits on are currently 30 minutes and 1 Gb of RAM.

• To use Abaqus interactively on SGI/Altix or Xeon/Linux type:
  abaqus [list of line command options, same as in Abaqus Manuals]
  Running Abaqus interactively on tun's access nodes is not allowed.
  You must first access a tun's compute node through:
  bsub -Is -n2 -W[hour]:minute tcsh
  Where -W[hour]:minute is a maximum wall clock access time to a tun's compute node.
  All users must be added to the access list for Abaqus. CAE is not supported by Abaqus (Sumulia Inc) on SGI/Altix.

Batch Jobs:

For large Abaqus jobs special scripts abaqus_batch is available to submit jobs to the batch queuing systems.

• To run Abaqus in batch on SGI/Altix or Xeon/Linux, type abaqus_batch

For more Computational Solid Mechanics Software Inquiries , please visit CSM-FAQs
If you have problems accessing Abaqus at NCSA, please send email to csm@ncsa.edu.

Abaqus 6.7 available on tungsten and cobalt :

Abaqus 6.7 is now installed on tungsten and cobalt, solver batch access only. For access type:

/usr/apps/csm/scripts/abaqus_batch671 (for batch access)

 

Abaqus in parallel runs:

The graph on the left side shows Abaqus scaling performance on NCSA's Origin 2000 for various parallel runs of 210,000 and 17,000 dofs fixed problem sizes using a built-in Abaqus procedure for rectangular beams under torsion loads.

Speedup (Sp) is defined as the wall clock time for a single processor run divided by the wall clock time using n# processors. It tells us how much performance is achieved by using multiple processors, or the benefit of solving the problem in parallel.

It can be seen that fairly good Speedup is achieved for the 210K dofs problem. The 17K dofs curve shows very limited Speedup even with 2 processors, while for 
16 processors a point of negative return is reached due to too much parallel overhead compared to the amount of computation. Speedup is limited when the problem size is too small. Therefore please don't use more than 4 processors unless your job has more than 200,000 dofs.

Speedup is always an increasing function of problem size and the more dofs one has, the more processors should be used to achieve optimum performance. Speedup is not a function of the size of time-dependant Abaqus procedures, and it completely depends on the number of dofs in your FE mesh domain.

The graph on the right side is showing Abaqus wall clock data for 1 million dof problem ran on SGI O2K, IBMp690, and HP SuperDome.

Scaling Performance on Origin 2000

Abaqus Wall Clock Data on Different Platforms