The Cray MTA, a multithreaded architecture, is a new parallel supercomputer installed at San Diego Supercomputer Center (SDSC). This machine has an architecture quite different from those of other contemporary parallel machines. It has a flat, shared memory without locality and has hardware support for very fine-grained multithreading. The machine and its associated parallelizing compiler promi...

A discussion of convergence acceleration techniques as they relate to computational fluid dynamics problems on unstructured meshes is given. Rather than providing a detailed description of particular methods, the various different building blocks of current solution techniques are discussed and examples of solution strategies using one or several of these ideas are given. Issues relating to uns...

A parallel method for the dynamic partitioning of unstructured meshes is described. The method introduces a new iterative optimisation technique known as relative gain optimisation which both balances the workload and attempts to minimise the interprocessor communications overhead. Experiments on a series of adaptively refined meshes indicate that the algorithm provides partitions of an equival...

Geometric relations in an irregular D polyhedron or tetrahedral mesh are often di cult to comprehend even for relatively few vertices A technique for illustrating such meshes which aids this comprehension is described in terms of several independent components i e edge representation viewpoint and perspective projection and light ing These images are suitable for embedding in dynamic displays o...

We present a framework to extract mesh features from unstructured two-manifold surfaces. Our method computes a collection of piecewise linear curves describing the salient features of surfaces, such as edges and ridge lines. We extend these basic techniques to a multiresolution setting which improves the quality of the results and accelerates the extraction process. The framework is semiautomat...

Many large-scale engineering and scientiic calculations involve repeated updating of variables on an unstructured mesh. To do these types of computations on distributed memory parallel computers, it is necessary to partition the mesh among the processors so that the load balance is maximized and inter-processor communication time is minimized. This can be approximated by the problem of partitio...

We propose a nonlinear manifold learning technique based on deep convolutional autoencoders that is appropriate for model order reduction of physical systems in complex geometries. Convolutional neural networks have proven to be highly advantageous compressing data arising from demonstrating slow-decaying Kolmogorov n-width. However, these are restricted structured meshes. Unstructured meshes o...