Bulletin of the American Physical Society
APS April Meeting 2015
Volume 60, Number 4
Saturday–Tuesday, April 11–14, 2015; Baltimore, Maryland
Session M5: Simulation Methods and Implementation |
Hide Abstracts |
Sponsoring Units: DCOMP Room: Key 1 |
Sunday, April 12, 2015 3:30PM - 3:42PM |
M5.00001: Simulating Isotope Enrichment by Gaseous Diffusion Cameron Reed A desktop-computer simulation of isotope enrichment by gaseous diffusion has been developed. The simulation incorporates two non-interacting point-mass species whose members pass through a cascade of cells containing porous membranes and retain constant speeds as they reflect off the walls of the cells and the spaces between holes in the membranes. A particular feature is periodic forward recycling of enriched material to cells further along the cascade along with simultaneous return of depleted material to preceding cells. The number of particles, the mass ratio, the initial fractional abundance of the lighter species, and the time between recycling operations can be chosen by the user. The simulation is simple enough to be understood on the basis of two-dimensional kinematics, and demonstrates that the fractional abundance of the lighter-isotope species increases along the cascade. The logic of the simulation will be described and results of some typical runs will be presented and discussed. [Preview Abstract] |
Sunday, April 12, 2015 3:42PM - 3:54PM |
M5.00002: Dynamic Adaptive Runtime Systems for Advanced Multipole Method-based Science Achievement Jackson DeBuhr, Matthew Anderson, Thomas Sterling, Bo Zhang Multipole methods are a key computational kernel for a large class of scientific applications spanning multiple disciplines. Yet many of these applications are strong scaling constrained when using conventional programming practices. Hardware parallelism continues to grow, emphasizing medium and fine-grained thread parallelism rather than the coarse-grained process parallelism favored by conventional programming practices. Emerging, dynamic task management execution models can go beyond these conventional practices to significantly improve both efficiency and scalability for algorithms like multipole methods which exhibit irregular and time-varying execution properties. We present a new scientific library, DASHMM, built on the ParalleX HPX-5 runtime system, which explores the use of dynamic adaptive runtime techniques to improve scalability and efficiency for multipole-method based scientific computing. DASHMM allows application scientists to rapidly create custom, scalable, and efficient multipole methods, especially targeting the Fast Multipole Method and the Barnes-Hut N-body algorithm. After a discussion of the system and its goals, some application examples will be presented. [Preview Abstract] |
Sunday, April 12, 2015 3:54PM - 4:06PM |
M5.00003: On verification and validation of spring fabric model Zheng Gao, Qiangqiang Shi, Yiyang Yang, Xiaolin Li An enhanced spring-mass model has been developed to mimic the complex behavior of parachute canopy in the air flow. Given the Young's modulus and Poisson's ratio, the model has the ability to duplicate the realistic strain and stress of the elastic membrane by including the angular deformation energy in the triangulated mesh. The numerical results verify the effectiveness of the proposed model and demonstrate its convergent property. In addition, GPU-based parallel computing techniques are applied to accelerate the computational speed and increase the resolution of numerical results. [Preview Abstract] |
Sunday, April 12, 2015 4:06PM - 4:18PM |
M5.00004: GEANT4 Hadron Monitor Simulation Studies for the Long-Baseline Neutrino Facility Timothy Watson, Amit Bashyal, Jaehoon Yu The hadron monitor to be incorporated into the beamline of the Long-Baseline Neutrino Facility at Fermilab is a crucial tool for alignment purposes as well as the correct functionality of the beam. Designing such a hadron monitor requires careful consideration of the challenges presented. The high-radiation environment from the considerably higher proton beam power expected for LBNF at the monitor location coupled with the need for relatively higher spatial resolution from the monitor will require an innovative new detector technology and design. To this end, computer simulations are a useful tool. Presented here are the results of hadron monitor design studies simulated using GEANT4. [Preview Abstract] |
Sunday, April 12, 2015 4:18PM - 4:30PM |
M5.00005: A Generalized 4th-Order Runge-Kutta Method for the Gross-Pitaevskii Equation Martin Kandes We present the implementation of a method-of-lines approach for numerically approximating solutions of the time-dependent Gross-Pitaevksii equation in non-uniformly rotating reference frames. Implemented in parallel using a hybrid MPI + OpenMP framework, which will allow for scalable, high-resolution numerical simulations, we utilize an explicit, generalized 4th-order Runge-Kutta time-integration scheme with 2nd- and 4th-order central differences to approximate the spatial derivatives in the equation. The principal objective of this project is to model the effect(s) of inertial forces on quantized vortices within weakly-interacting dilute atomic gas Bose-Einstein condensates in the mean-field limit of the Gross-Pitaevskii equation. Here, we discuss our work-to-date and preliminary results. [Preview Abstract] |
Sunday, April 12, 2015 4:30PM - 4:42PM |
M5.00006: Fast Multipole Method for Coulomb Interaction Based on Traceless Totally Symmetric Tensor He Huang, Rui Li, Jie Chen, Li-Shi Luo, He Zhang The fast multipole method (FMM) is widely used to calculate the Coulomb interaction between a huge amount of charged particles. The efficiency of FMM scales with $O(N)$ for $N$ particles with any arbitrary distribution. Hence it is apposite for problems with complicated charge distribution or geometry. Under the same FMM framework, there are different approaches, such as using spherical harmonic functions or Maxwell Cartesian tensors. Here we will present a version using traceless totally symmetric Maxwell Cartesian tensor. The previous Maxwell Cartesian tensor based FMM uses totally symmetric tensor. There are $(n+1)(n+2)/2$ independent elements in a rank $n$ totally symmetric tensor. However, there are only $2n+1$ independent elements in a rank $n$ traceless totally symmetric tensor, due to which the efficiency of the traceless version is highly improved compared with the old version, especially when high accuracy is needed and high rank tensors are used. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700