Fractional factorial design for parameter sweep experiments using Nimrod/E

  • Authors:

  • T. C. Peachey;N. T. Diamond;D. A. Abramson;W. Sudholt;A. Michailova;S. Amirriazi

  • Affiliations:

  • Monash University, Clayton, 3800, VIC, Australia;Monash University, Clayton, 3800, VIC, Australia;((Correspd. E-mail: davida@csse.monash.edu.au) Monash University, Clayton, 3800, VIC, Australia;Institute of Organic Chemistry, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland;Department of Bioengineering, University of California San Diego, and San Diego Supercomputer Centre, 9500 Gilman Drive, La Jolla, CA, USA;Department of Bioengineering, University of California San Diego, and San Diego Supercomputer Centre, 9500 Gilman Drive, La Jolla, CA, USA

  • Venue:
  • Scientific Programming - Large-Scale Programming Tools and Environments
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

The techniques of formal experimental design and analysis are powerful tools for scientists and engineers. However, these techniques are currently underused for experiments conducted with computer models. This has motivated the incorporation of experimental design functionality into the Nimrod tool chain. Nimrod has been extensively used for exploration of the response of models to their input parameters; the addition of experimental design tools will combine the efficiency of carefully designed experiments with the power of distributed execution. This paper describes the incorporation of one type of design, the fractional factorial design, and associated analysis tools, into the Nimrod framework. The result provides a convenient environment that automates the design of an experiment, the execution of the jobs on a computational grid and the return of results, and which assists in the interpretation of those results. Several case studies are included which demonstrate various aspects of this approach.