3:15 PM-5:15 PM
Law School, Room 290
Unstructured grids are increasingly being used in many areas of scientific computing because of the flexibility in adapting to complex geometries and solution features. However, the lack of natural coarse structures also makes it difficult to implement domain decomposition and multigrid type algorithms efficiently, especially on high performance parallel computers. In this two part minisymposium, we bring together experts in this field to discuss both algorithmic and theoretical aspects, as well as applications to CFD and structures.
Domain decomposition (DD) and multi-grid (MG) algorithms have demonstrated their potential in solving challenging problems in diverse areas. However, when the topology and geometry of the grid is irregular, conventional implementations of these methods often have difficulties in delivering optimal performance. In addition, implementational issues arise when using high performance computing architectures. To our knowledge, this direction of research has the potential to greatly impact the efficiency of industrial analysis codes in CFD, structures, and related areas, yet there are significant theoretical and implementational issues that must be resolved.
This two-part minisymposium brings together experts actively working on the theory and application of unstructured grids in solving real world applications.
Organizers: Tony F. Chan, University of California, Los Angeles; Timothy J. Barth and Wei-Pai Tang, NASA Ames Research Center
AN97 Homepage | Program Updates|
Registration | Hotel and Dormitory Information | Transportation | Program-at-a-Glance | Program Overview