</a>Polyhedral design of computer memory utilization processes

Polyhedral Empowerment of Networks through Symmetry: psycho-social implications for organization and global governance (Part #10)

[Parts: First | Prev | Next | Last | All] [Links: To-K | From-K | From-Kx | Refs ]

The power of supercomputers is partly due to their use of a design based on a hypercube configuration of distributed memory parallel computers (see N-dimensional modified hypercube). The plurality of nodes or cells is interconnected to provide a shared memory with processors of the network and their memory providing the network routing and shared memory. Distributed memory parallel computers offer both the potential for a dramatic improvement in cost/performance over conventional supercomputers.

Of relevance here is the application to resource intensive computations in fluid dyanmics required by applications, for example:

• COMET: incorporating mathematical models of a wide range of thermo-fluids and solids phenomena, powerful solvers enabling full complex-geometry capabilities, highly efficient and stable set of numerical solution algorithms, flexible pre-processing and post-processing facilities. With no restriction on the shape of cells, grids are automatically optimized for complex geometries to achieve maximum accuracy with minimum effort. When compared with simple tetrahedral meshes, polyhedral meshes offer the same accuracy with typically half the memory requirement and a quarter of the computing time.
• STAR-CD: a simulation tool for computational continuum mechanics and computational fluid dynamics. he combination of high quality automatic polyhedral mesh generation and efficient solver technology provides improved ease-of-use, faster turnaround times and increased solution accuracy, for all classes of flow, in all application areas.

Combinatorial computational geometry: The primary goal of research in combinatorial computational geometry is to develop efficient algorithms and data structures for solving problems stated in terms of basic geometrical objects: points, line segments, polygons, polyhedra, etc.

There is tremendous pressure to develop computers of higher performance in response to certain modelling challenges, notably those related to climate and ocean currents and to challenges of astrophysics and fundamental physics. But it is curious that the applications of relevance to psycho-social challenges have been limited to relatively simplistic forms of "global modelling". There has been no implication that there might be equally demanding applications, of equivalent significance for humanity, associated with building richer and more complex psycho-social systems.

[Parts: First | Prev | Next | Last | All] [Links: To-K | From-K | From-Kx | Refs ]