SIMULTANEOUS OPTIMISATION: STRATEGIES FOR USING PARALLELIZATION EFFICENTLY
Editor: Christian Weber, Stephan Husung, Marco CantaMESsa, Gaetano Cascini, Dorian Marjanovic, Serena Graziosi
Author: Wuensch, Andreas; Jordan, André; Vajna, Sándor
Institution: Otto-von-Guericke University Magdeburg, Germany
Section: Design Methods and Tools - part 2
Efficiency plays a major role in any facet of product development. The product has to be efficient itself, but the processes of product development have to be efficient as well. Analysis and simulation enable engineers to evaluate new products without the need for physical prototyping and improve them by optimisation methods. Since parallel computing was introduced the process of product optimisation became more efficient. However, it is very susceptible to bottlenecks like any the lack of available resources.
We transfer the approach of simultaneous engineering to an optimisation framework. The term simultaneous optimisation is introduced in order to use the available resources in an efficient way and to reduce idleness of the resources. To verify the framework it was tested in an industry-related workstation cluster. In this approach resources are floating and can become available or busy in order to the current use of a machine or the unavailability of licenses. The presented approach works in both homogenous and heterogeneous workstation clusters.