A grid-enabled problem solving environment for parallel computational engineering design

Abstract

This paper describes the development and application of a piece of engineering software that provides a Problem Solving Environment (PSE) capable of launching, and interfacing with, computational jobs executing on remote resources on a computational Grid. In particular it is demonstrated how a complex, serial, engineering optimisation code may be efficiently parallelised, Grid-enabled and embedded within a PSE. The environment is highly flexible, allowing remote users from different sites to collaborate, and permitting computational tasks to be executed in parallel across multiple Grid resources, each of which may be a parallel architecture. A full working prototype has been built and successfully applied to a computationally demanding engineering optimisation problem. This particular problem stems from elastohydrodynamic lubrication and involves optimising the computational model for a lubricant based on the match between simulation results and experimentally observed data.

Using adjoint error estimation techniques for elastohydrodynamic lubrication line contact problems

“…the matrix K in (12) depending upon the quadrature rule that is selected for the evaluation of (2). In this work the following standard equation is used:…”

“…The purpose of separating the bodies with a thin lubricating film is to reduce the friction in the contact, thus increasing the efficiency of energy transfer through the contact, and reducing the wear on the contacting elements. Consequently, when modelling an EHL contact numerically, while flow field properties such as the pressure distribution or the thickness of the film are significant, it is often quantities such as the friction, that are of most interest [1,2].…”

Reliable performance prediction for multigrid software on distributed memory systems

“…[3,[11][12][13]19]. This is typically based upon using a geometric partition of the coarsest mesh that is present and then mapping this in the natural manner to each of the finer meshes.…”

Mathematical Modelling of Chemical Diffusion through Skin using Grid-based PSEs

“…Many providers of Grid resources are actively limiting the number of submissions per person able to be executed simultaneously, and hence running large numbers concurrently is often not possible on the chosen resource. In the work of Goodyer et al [11] it was seen how the use of a parallel environment was beneficial for solution time of a previously serial optimisation application, through concurrent simulation of independent cases with similar parameter sets. Here we extend this idea to take the multiple independent simulations inside one large MPI job.…”

“…In many of these cases, the problems being tackled on these systems are not so large that parallel computing is being used for single cases, but instead, multiple production runs, sometimes within the context of a design optimization process, are run on a multiprocessor architecture, e.g. [11].…”