Publications

Abstract not provided.

As current experimental and simulation methods cannot determine the mobility of flat boundaries across the large misorientation phase space, we have developed a computational method for imposing an artificial driving force on boundaries. In a molecular dynamics simulation, this allows us to go beyond the inherent timescale restrictions of the technique and induce non-negligible motion in flat boundaries of arbitrary misorientation. For different series of symmetric boundaries, we find both expected and unexpected results. In general, mobility increases as the grain boundary plane deviates from (111), but high-coincidence and low-angle boundaries represent special cases. These results agree with and enrich experimental observations.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Abstract not provided.

This paper is about making reversible logic a reality for supercomputing. Reversible logic offers a way to exceed certain basic limits on the performance of computers, yet a powerful case will have to be made to justify its substantial development expense. This paper explores the limits of current, irreversible logic for supercomputers, thus forming a threshold above which reversible logic is the only solution. Problems above this threshold are discussed, with the science and mitigation of global warming being discussed in detail. To further develop the idea of using reversible logic in supercomputing, a design for a 1 Zettaflops supercomputer as required for addressing global climate warming is presented. However, to create such a design requires deviations from the mainstream of both the software for climate simulation and research directions of reversible logic. These deviations provide direction on how to make reversible logic practical.

In contrast to traditional terascale simulations that have known, fixed data inputs, dynamic data-driven (DDD) applications are characterized by unknown data and informed by dynamic observations. DDD simulations give rise to inverse problems of determining unknown data from sparse observations. The main difficulty is that the optimality system is a boundary value problem in 4D space-time, even though the forward simulation is an initial value problem. We construct special-purpose parallel multigrid algorithms that exploit the spectral structure of the inverse operator. Experiments on problems of localizing airborne contaminant release from sparse observations in a regional atmospheric transport model demonstrate that 17-million-parameter inversion can be effected at a cost of just 18 forward simulations with high parallel efficiency. On 1024 Alphaserver EV68 processors, the turnaround time is just 29 minutes. Moreover, inverse problems with 135 million parameters - corresponding to 139 billion total space-time unknowns - are solved in less than 5 hours on the same number of processors. These results suggest that ultra-high resolution data-driven inversion can be carried out sufficiently rapidly for simulation-based 'real-time' hazard assessment.

Abstract not provided.

Abstract not provided.

We address the problem of partitioning and dynamic load balancing on clusters with heterogeneous hardware resources. We propose DRUM, a model that encapsulates hardware resources and their interconnection topology. DRUM provides monitoring facilities for dynamic evaluation of communication, memory, and processing capabilities. Heterogeneity is quantified by merging the information from the monitors to produce a scalar number called 'power.' This power allows DRUM to be used easily by existing load-balancing procedures such as those in the Zoltan Toolkit while placing minimal burden on application programmers. We demonstrate the use of DRUM to guide load balancing in the adaptive solution of a Laplace equation on a heterogeneous cluster. We observed a significant reduction in execution time compared to traditional methods.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Combinatorial algorithms have long played a pivotal enabling role in many applications of parallel computing. Graph algorithms in particular arise in load balancing, scheduling, mapping and many other aspects of the parallelization of irregular applications. These are still active research areas, mostly due to evolving computational techniques and rapidly changing computational platforms. But the relationship between parallel computing and discrete algorithms is much richer than the mere use of graph algorithms to support the parallelization of traditional scientific computations. Important, emerging areas of science are fundamentally discrete, and they are increasingly reliant on the power of parallel computing. Examples include computational biology, scientific data mining, and network analysis. These applications are changing the relationship between discrete algorithms and parallel computing. In addition to their traditional role as enablers of high performance, combinatorial algorithms are now customers for parallel computing. New parallelization techniques for combinatorial algorithms need to be developed to support these nontraditional scientific approaches. This chapter will describe some of the many areas of intersection between discrete algorithms and parallel scientific computing. Due to space limitations, this chapter is not a comprehensive survey, but rather an introduction to a diverse set of techniques and applications with a particular emphasis on work presented at the Eleventh SIAM Conference on Parallel Processing for Scientific Computing. Some topics highly relevant to this chapter (e.g. load balancing) are addressed elsewhere in this book, and so we will not discuss them here.

Abstract not provided.

In this article we describe stress nets, a technique for exploring 2D tensor fields. Our method allows a user to examine simultaneously the tensors eigenvectors (both major and minor) as well as scalar-valued tensor invariants. By avoiding noise-advection techniques, we are able to display both principal directions of the tensor field as well as the derived scalars without cluttering the display. We present a CPU-only implementation of stress nets as well as a hybrid CPU/GPU approach and discuss the relative strengths and weaknesses of each. Stress nets have been used as part of an investigation into crack propagation. They were used to display the directions of maximum shear in a slab of material under tension as well as the magnitude of the shear forces acting on each point. Our methods allowed users to find new features in the data that were not visible on standard plots of tensor invariants. These features disagree with commonly accepted analytical crack propagation solutions and have sparked renewed investigation. Though developed for a materials mechanics problem, our method applies equally well to any 2D tensor field having unique characteristic directions.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Empirical studies suggest that consumption is more sensitive to current income than suggested under the permanent income hypothesis, which raises questions regarding expectations for future income, risk aversion, and the role of economic confidence measures. This report surveys a body of fundamental economic literature as well as burgeoning computational modeling methods to support efforts to better anticipate cascading economic responses to terrorist threats and attacks. This is a three part survey to support the incorporation of models of economic confidence into agent-based microeconomic simulations. We first review broad underlying economic principles related to this topic. We then review the economic principle of confidence and related empirical studies. Finally, we provide a brief survey of efforts and publications related to agent-based economic simulation.

Abstract not provided.

Simulations within density functional theory (DFT) are a common component of research into the physics of materials. With the broad success of DFT, it is easily forgotten that computational DFT methods invariably do not directly represent simulated properties, but require careful construction of models that are computable approximations to a physical property. Perhaps foremost among these computational considerations is the routine use of the supercell approximation to construct finite models to represent infinite systems. Pitfalls in using supercells (k-space sampling, boundary conditions, cell sizes) are often underappreciated. We present examples (e.g. vacancy defects) that exhibit a surprising or significant dependence on supercells, and describe workable solutions. We describe procedures needed to construct meaningful models for simulations of real material systems, focusing on k-space and cell size issues.

Abstract not provided.

Abstract not provided.