Showing papers by "Steven J. Plimpton published in 2005"
TL;DR: The implementation of a recently proposed parallel algorithm that finds strongly connected components in distributed graphs, and how it is used in a radiation transport solver is described.
81 citations
TL;DR: This work presents an asynchronous message-passing algorithm that performs sweeps simultaneously in multiple ordinate directions, a simple geometric heuristic to prioritize the computational tasks that a processor works on, and an algorithm for detecting and eliminating cycles that sometimes exist in unstructured grids and can prevent sweeps from successfully completing.
Abstract: The method of discrete ordinates is commonly used to solve the Boltzmann transport equation. The solution in each ordinate direction is most efficiently computed by sweeping the radiation flux across the computational grid. For unstructured grids this poses many challenges, particularly when implemented on distributed-memory parallel machines where the grid geometry is spread across processors. We present several algorithms relevant to this approach: (a) an asynchronous message-passing algorithm that performs sweeps simultaneously in multiple ordinate directions, (b) a simple geometric heuristic to prioritize the computational tasks that a processor works on, (c) a partitioning algorithm that creates columnar-style decompositions for unstructured grids, and (d) an algorithm for detecting and eliminating cycles that sometimes exist in unstructured grids and can prevent sweeps from successfully completing. Algorithms (a) and (d) are fully parallel; algorithms (b) and (c) can be used in conjunction with (a) to achieve higher parallel efficiencies. We describe our message-passing implementations of these algorithms within a radiation transport package. Performance and scalability results are given for unstructured grids with up to 3 million elements (500 million unknowns) running on thousands of processors of Sandia National Laboratories' Intel Tflops machine and DEC-Alpha CPlant cluster.
39 citations
TL;DR: In this article, the effect of deformation sequence and history on the inelastic behavior of copper interfaces on the nanoscale was investigated using an asymmetric 45 deg tilt bicrystal interface.
Abstract: Molecular dynamics calculations are performed to study the effect of deformation sequence and history on the inelastic behavior of copper interfaces on the nanoscale. An asymmetric 45 deg tilt bicrystal interface is examined, representing an idealized high-angle grain boundary interface. The interface model is subjected to three different deformation paths: tension then shear, shear then tension, and combined proportional tension and shear. Analysis shows that path-history dependent material behavior is confined within a finite layer of deformation around the bicrystal interface. The relationships between length scale and interface properties, such as the thickness of the path-history dependent layer and the interface strength, are discussed in detail.
13 citations
01 Dec 2005
TL;DR: In this paper, a level set based feature-scale model for low-pressure and plasma enhanced chemical vapor deposition (LPCVD, LPCVD and PECVD) is presented.
Abstract: The surface micromachining processes used to manufacture MEMS devices and integrated circuits transpire at such small length scales and are sufficiently complex that a theoretical analysis of them is particularly inviting. Under development at Sandia National Laboratories (SNL) is Chemically Induced Surface Evolution with Level Sets (ChISELS), a level-set based feature-scale modeler of such processes. The theoretical models used, a description of the software and some example results are presented here. The focus to date has been of low-pressure and plasma enhanced chemical vapor deposition (low-pressure chemical vapor deposition, LPCVD and PECVD) processes. Both are employed in SNLs SUMMiT V technology. Examples of step coverage of SiO2 into a trench by each of the LPCVD and PECVD process are presented.
8 citations
06 Jun 2005
TL;DR: In this article, a rendezvous technique is used to search for elements in one grid that contain nodal points of the other, which has the advantage of enabling the grid transfer operation to be load-balanced separately from the remainder of the computations.
Abstract: A number of computational procedures employ multiple grids on which solutions are computed. For example, in multiphysics simulations a primary grid may be used to compute mechanical deformation of an object while a secondary grid is used for thermal conduction calculations. When modeling coupled thermo-mechanical effects, solution data must be interpolated back and forth between the grids each timestep. On a parallel machine, this grid transfer operation can be challenging if the two grids are decomposed across processors differently for reasons of computational efficiency. If the grids move or adapt separately, the complexity of the operation is compounded. In this paper, we describe two grid transfer algorithms suitable for massively parallel simulations which use multiple grids. They use a rendezvous technique wherein a third decomposition is used to search for elements in one grid that contain nodal points of the other. This has the advantage of enabling the grid transfer operation to be load-balanced separately from the remainder of the computations. The algorithms are designed for use within the multi-physics code SIERRA, an object-oriented framework developed at Sandia. Performance and scalability results are given for the grid transfer operation running on up to 1024 processors of two large parallel machines, the Intel Tflops (ASCI Red) and DEC-Alpha CPlant cluster.