Dakota

Overview

Dakota is a program to perform advanced parametric analyses to enable design exploration, model calibration, risk analysis, and quantification of margins and uncertainty with computational models.

Versions 6.10.0 and 6.15.0 are installed on the CSF. Both are “binary” installs – i.e., not compiled from source.

Restrictions on use

There are no restrictions on accessing this software on the CSF. Please review the various licenses under which the software is released and ensure your usage falls within those licenses. In particular, the GUI is released under the Eclipse Public License and the main app is released under the GNU Lesser General Public License.

Set up procedure

We now recommend loading modulefiles within your jobscript so that you have a full record of how the job was run. See the example jobscript below for how to do this. Alternatively, you may load modulefiles on the login node and let the job inherit these settings.

Load one of the following modulefiles:

module load apps/binapps/dakota/6.15.0       # GUI and command-line versions available
module load apps/binapps/dakota/6.10.0       # GUI and command-line versions available

Running the application

Please do not run Dakota on the login node. Jobs should be submitted to the compute nodes via batch.

Serial batch job submission

Create a batch submission script (which will load the modulefile in the jobscript), for example:

#!/bin/bash --login
#$ -cwd             # Job will run from the current directory
                    # NO -V line - we load modulefiles in the jobscript

module load apps/binapps/dakota/6.15.0

dakota -input inputfile args ...

Submit the jobscript using:

qsub scriptname

where scriptname is the name of your jobscript.

Parallel batch job submission

It is recommended that you consult the Dakota manuals for details on how to perform parallel analysis using Dakota.

Parallel execution of Dakota is still being confirmed.

where scriptname is the name of your jobscript.

Interactive GUI usage

The Dakota GUI should not be started on the login nodes. It is expected that the GUI be used just to prepare models and visualize results. But the main dakota process jobs should be submitted to batch, without use of the GUI (see above.)

To run the GUI, first start an interactive session:

qrsh -l short -pe smp.pe 4
  #
  # Wait until you've been logged in to a compute node.

module load apps/binapps/dakota/6.15.0
dakota-ui.sh

# When you have finished with the GUI, return to the login node.
exit

Further info

• Various manuals available on the Dakota website.
• Dakota website.

Updates

None.