A5034862374- Parallel Computing and Optimization Techniques
- Advanced Data Storage Technologies
- Nuclear and radioactivity studies
- Nuclear reactor physics and engineering
- Distributed and Parallel Computing Systems
- nanoparticles nucleation surface interactions
- Medical Image Segmentation Techniques
- Image Retrieval and Classification Techniques
- Microstructure and mechanical properties
- Graphite, nuclear technology, radiation studies
- Advanced Image and Video Retrieval Techniques
- Polymer crystallization and properties
- Metal and Thin Film Mechanics
- Industrial Vision Systems and Defect Detection
- Digital Transformation in Industry
- Ion-surface interactions and analysis
- Nuclear Physics and Applications
- Electrostatics and Colloid Interactions
- Coagulation and Flocculation Studies
- Advanced Vision and Imaging
- Advanced Chemical Physics Studies
- Image and Object Detection Techniques
- Remote-Sensing Image Classification
- Polymer Foaming and Composites
- Nanotechnology research and applications
Los Alamos National Laboratory
2008-2025
CCS Associates (United States)
2022
Computational Physics (United States)
2009
University of Michigan
1994-1997
After several years of effort involved in merging the Los Alamos National Laboratory MCNP5 and MCNPX codes, 2013 first production release version 6 Monte Carlo N-Particle ® , or MCNP code MCNP6.1 was distributed publicly. Since then, three significant releases have been issued: MCNP6.1.1beta 2014, MCNP6.2 2018, MCNP6.3 2023. While each always contains new features, enhancements, bug fixes, has had a different primary focus, ranging from improved calculational efficiency to powerful utilities...
On the path to exascale landscape of computer device architectures and corresponding programming models has become much more diverse. While various low-level performance portable are available, support at application level lacks behind. To address this issue, we present block-structured adaptive mesh refinement (AMR) framework Parthenon, derived from well-tested widely used Athena++ astrophysical magnetohydrodynamics code, but generalized serve as foundation for a variety downstream...
Abstract In part I of this paper, we presented two efficient front‐tracking methods to simulate the growth a spherulite within an imposed temperature field. second present method that predicts final microstructure in macroscopic by coupling these techniques with (a) stochastic model for nucleation individual spherulites, (b) cellular impingement and solid fraction evolution (c) Finite Difference Method (FDM) latent heat release diffusion. The tracks physical phenomena on several length...
We present RaveGrid, a software that efficiently converts raster image to scalable vector comprised of polygons whose boundaries conform the edges in image. The resulting has good visual quality and fidelity can be displayed at various sizes on display screen resolutions. render images SVG (scalable graphics) format or EPS (Encapsulated Postscript). ubiquity data graphics, Web, communications, as well wide range devices, from big TVs hand-held cellular phones support display, calls for more...
Abstract In this two‐part paper, we present a model for the solidification of semicrystalline thermoplastic polymers that links microscopic and macroscopic length scales. The accounts important physical phenomena occurring during spherulitic cyrstallization predicts microstructural evolution. first part, concentrate on growth individual spherulites. We fast accurate shape spherulite as well path its constitutive lamellae in any thermal situation. part II paper shall couple new with...
We present timing and performance numbers for a short-range parallel molecular dynamics (MD) code, SPaSM, that has been rewritten the heterogeneous Roadrunner supercomputer. Each compute node consists of two AMD Opteron dualcore microprocessors four PowerXCell 8i enhanced Cell microprocessors, so there are MPI ranks per node, each with one Cell. The interatomic forces computed on Cells (each PPU eight SPU cores), while Opterons used to direct inter-rank communication perform I/O-heavy...
Abstract We describe the implementation of a short‐range parallel molecular dynamics (MD) code, SPaSM, on heterogeneous general‐purpose Roadrunner supercomputer. Each ‘TriBlade’ compute node consists two AMD Opteron dual‐core microprocessors and four IBM PowerXCell 8i enhanced Cell (each consisting one PPU eight SPU cores), so that there are MPI ranks per node, each with Cell. will briefly architecture some initial hybrid programming approaches have been taken, focusing SPaSM application as...
Neuroscience has revealed many properties of neurons and the functional organization visual cortex that are believed to be essential human vision, but missing in standard artificial neural networks. Equally important may sheer scale requiring ~1 petaflop computation, while experience greatly exceeds computer vision datasets: retina delivers petapixel/year brain, driving learning at levels cortical system. We describe work Los Alamos National Laboratory (LANL) develop large-scale models on...
We present timing and performance numbers for a short-range parallel molecular dynamics (MD) code, SPaSM, that has been rewritten the heterogeneous Roadrunner supercomputer. Each compute node consists of two AMD Opteron dualcore microprocessors four PowerXCell 8i enhanced Cell microprocessors, so there are MPI ranks per node, each with one Cell. The interatomic forces computed on Cells (each PPU eight SPU cores), while Opterons used to direct inter-rank communication perform I/O-heavy...
We present a simple cascading algorithm for rapid hierarchical image segmentation based on perceptually driven contour completion and scene statistics. start with an initial fine-scale of obtained by perceptual partial contours into polygonal regions using region-contour correspondences established Delaunay triangulation edge pixels. The resulting polygon size distribution is analyzed dominant mode granularity the image. Polygons whose sizes are less than or equal to this merged their...