Distributed control parallelism in multidisciplinary aircraft design

Multidisciplinary design optimization (MDO) for large-scale engineering pro blems poses many challenges (e,g, the design of an efficient concurrent par adigm for global optimization based on disciplinary analyses, expensive com putations over vast data sets, etc.). This work focuses on the application of distributed schemes for massively parallel architectures to MDO problems , as a tool for reducing computation time and solving larger problems. The specific problem considered here is configuration optimization of a high sp eed civil transport (HSCT), and the efficient parallelization of the embedd ed paradigm for reasonable design space identification. Two distributed dyn amic load balancing techniques (random polling and global round robin with message combining) and two necessary termination detection schemes (global task count and token passing) were implemented and evaluated in terms of ef fectiveness and scalability to large problem sizes and a thousand processor s. The effect of certain parameters on execution time was also inspected. E mpirical results demonstrated stable performance and effectiveness for all schemes, and the parametric study showed that the selected algorithmic para meters have a negligible effect on performance. Copyright (C) 1999 John Wil ey & Sons, Ltd.