Advertisement

The Development of a Grid-based Engineering Design Problem Solving Environment.

  • A. D. Scurr
  • A. J. Keane
Conference paper

Abstract

This paper gives an overview of the grid based Engineering Design Problem Solving Environment (PSE) being developed at Southampton University. Following some background comments on PSE definition and function, the conceptual architecture of a PSE tool developed jointly with Cardiff University is shown. This has been integrated with the Engineering Design PSE using the Cardiff VCCE and XML component model. Essentially, VCCE forms a GUI front-end to our PSE and provides the ability to setup and execute a computation by creating a task graph from available components via drag and drop operations on a sketchpad display. A detailed description of the PSE component structure is presented next, with particular reference to the form and type of communication required between components. To overcome some communications restrictions and to provide the framework for a generic PSE system, a more general communications sub-system is being developed, based on a message passing design, using CORBA calls for component communication within an intranet cluster and using XML/SOAP for component communication across intranets.

As developed, the current PSE will be able to compute single or multi-point design evaluations over network clusters. However, in order to fully exploit parallel computation over intranet or even internet clusters and to provide an environment which more naturally meets the data-centric view of users, two additional major enhancements to the PSE are planned. The first concerns scheduling and task farming. The ultimate goal is to achieve within the PSE, an asynchronous computational workflow pattern where analysis tasks can seek to exploit whatever computational resources are available in various workstation clusters (i.e., task farming), running independently of search/optimisation tasks. The second enhancement concerns computational resource control and job control and the setting up of an Engineering Design Grid Portal. Some consideration is also given to the need to move towards a full server-side component model, particularly when the PSE has to cope with multiuser access to distributed databases.

Keywords

Design Point Analysis Task Response Surface Modeling Task Graph Client Application 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    A.J. Keane, The Options Design Exploration System Reference Manual and User Guide — Version B3.0 February 2000.Google Scholar
  2. [2]
    O. F. Rana, M. Li, M. S. Shields, and D. W. Walker, A Wrapper Generator for Wrapping High Performance Legacy Codes as Java/CORBA Components in Proceedings of the IEEE/ACM SC2000 Conference, held in Dallas, TX, Nov 10-12, 2000.Google Scholar
  3. [3]
    Myers, R. H. and Montgomery, D. C. (1995), Response Surface Methodology: Process and product optimization using designed experiments, John Wiley and Sons inc.Google Scholar
  4. [4]
    Jones, D. R, Schonlau, M. and Welch, W. J. (1998), Efficient global optimization of expensive black-box functions, Journal of Global Optimization, 13,455–492.MathSciNetMATHCrossRefGoogle Scholar
  5. [5]
    G. von Laszewski, Ian Foster, Jarek Gawor, Peter Lane, Mike Russell. Designing Grid-based Problem Solving Environments and Portals. HICSS 2001.Google Scholar
  6. [6]
    The Globus Project web-site: <http://www.globus.org/>Google Scholar
  7. [7]
    The EZ-Grid Project web-site: <http://www.cs.uh.edu/-ezgrid/EZ-Grid.htm>Google Scholar

Copyright information

© Springer-Verlag London 2002

Authors and Affiliations

  1. 1.School of Engineering Sciences University of SouthamptonUK

Personalised recommendations