Bulletin of the American Physical Society
2013 Joint Meeting of the APS Division of Atomic, Molecular & Optical Physics and the CAP Division of Atomic, Molecular & Optical Physics, Canada
Volume 58, Number 6
Monday–Friday, June 3–7, 2013; Quebec City, Canada
Session M5: Invited Session: Petascale Computing and Beyond: Applications and Opportunities for Atomic, Molecular, and Chemical Dynamics |
Hide Abstracts |
Chair: Phillip Stancil, University of Georgia Room: 301 |
Thursday, June 6, 2013 8:00AM - 8:30AM |
M5.00001: From Petascale to Exascale: Prospects for Transforming Atomic, Molecular, and Chemical Dynamics with Leadership Computing Invited Speaker: Jack Wells Modeling and simulation with petascale computing has supercharged the process of innovation and understanding, dramatically accelerating time-to-insight and time-to-discovery. From petascale modeling of combustion for advanced engines, to designing bio-inspired catalysts for renewable energy, to exploring the evolution of complex systems such as our earth's climate, or breakthroughs gained from quantum many-body applications in chemical and nuclear physics, petascale computing is delivering high impact results that are transforming science and engineering. This presentation will provide an overview of the unique computational resources and user programs at the Oak Ridge Leadership Computing Facility (OLCF) at DOE's Oak Ridge National Laboratory, discuss a range of ambitious computational research projects underway in atomic, molecular, and chemical physics, and discuss scientific opportunities and challenges associated with advancing computational capabilities to the exascale. [Preview Abstract] |
Thursday, June 6, 2013 8:30AM - 9:00AM |
M5.00002: Electron-impact calculations of near-neutral atomic systems utilising Petascale computer architectures Invited Speaker: Connor Ballance Over the last couple of decades, a number of advanced non-perturbative approaches such as the R-matrix[1,2,3], TDCC[4] and CCC[5] methods have made great strides in terms of improved target representation and investigating fundamental 2-4 electron problems. However, for the electron-impact excitation of near-neutral species or complicated open-shell atomic systems we are forced to make certain compromises in terms of the atomic structure and/or the number of channels included in close-coupling expansion of the subsequent scattering calculation. The availability of modern supercomputing architectures with hundreds of thousands of cores, and the emergence new opportunities through GPU usauge offers one possibility to address some of these issues. To effectively harness this computational power will require significant revision of the existing code structures. I shall discuss some effective strategies within a non-relativistic and relativistic R-matrix framework using the examples detailed below. The goal is to extend existing R-matrix methods from 1-2 thousand close coupled channels to 10,000 channels. With the construction of the ITER[6] experiment in Cadarache, which will have Tungsten plasma-facing components, there is an urgent diagnostic need for the collisional rates for the near-neutral ion stages. In particular, spectroscopic diagnostics of impurity influx require accurate electron-impact excitation and ionisation as well as a good target representation. There have been only a few non-perturbative collisional calculations for this system, and the open-f shell ion stages provide a daunting challenge even for perturbative approaches [7]. I shall present non-perturbative results for for the excitation and ionisation of W$^{3+}$ and illustrate how these fundamental calculations can be integrated into a meaningful diagnostic for the ITER device.\\[4pt] [1] P G Burke, V M Burke, K M Dunseath, J. Phys. B At. Mol. Opt. Phys. 27, 5341 (1994) \newline [2] O. Zatsarinny, Comput. Phys. Commun. 174, 273 (2006) \newline [3] C.P. Ballance D.C. Griffin, J. Phys. B At. Mol. Opt. Phys. 37, 2943 (2004) \newline [4] M S Pindzola, F Robicheaux, S D Loch, J C Berengut, J Colgan, M Foster, D C Griffin, C P Ballance, N R Badnell, M C Witthoeft, D R Schultz and T Minami, J. Phys.: Conf. Ser. 88 012012 doi:10.1088/1742-6596/88/1/012012 (2007) \newline [5] I Bray, D V Fursa, A S Kheifets and A T Stelbovics, J. Phys. B: At. Mol. Opt. Phys. 35 R117 doi:10.1088/0953-4075/35/15/201 (2002) \newline [6] http://www.iter.org/ \newline [7] N. R. Badnell, C. P. Ballance, D. C. Griffin, and M. O'Mullane, Phys. Rev. A 85, 052716 (2012) [Preview Abstract] |
Thursday, June 6, 2013 9:00AM - 9:30AM |
M5.00003: Large-scale Quantum Calculations of Inelastic and Reactive Scattering in Cold and Ultracold Gases Invited Speaker: Balakrishnan Naduvalath Recent advances in cooling and trapping atomic and molecular systems have led to exciting opportunities for the investigation of inelastic and reactive collisions in ultra cold gases. This opens up novel possibilities for the control and manipulation of chemical reaction dynamics and molecular scattering events in the extreme quantum regime. Quantum coupled-channel calculations have become the preferred approach to describe these events theoretically. While such calculations can be performed routinely for triatomic systems composed of light atoms, they become computationally intractable as the size and dimensionality of the system increase. Even for triatomic systems, calculations become prohibitive when all internal quantum numbers, including, rotation, vibration, spin and hyperfine levels are considered. I will discuss recent progress in inelastic and reactive collisions of atom-diatom and diatom-diatom systems at cold and ultra cold temperatures and delineate challenges involved in moving beyond three and four atomic systems. I will further discuss how these problems could be tackled with emerging advances in CPU/GPU architectures and availability of petascale computational platforms. [Preview Abstract] |
Thursday, June 6, 2013 9:30AM - 10:00AM |
M5.00004: Towards the simulation of molecular collisions with a superconducting quantum computer Invited Speaker: Michael Geller I will discuss the prospects for the use of large-scale, error-corrected quantum computers to simulate complex quantum dynamics such as molecular collisions. This will likely require millions qubits. I will also discuss an alternative approach [M. R. Geller et al., arXiv:1210.5260] that is ideally suited for today's superconducting circuits, which uses the single-excitation subspace (SES) of a system of n tunably coupled qubits. The SES method allows many operations in the unitary group SU(n) to be implemented in a single step, bypassing the need for elementary gates, thereby making large computations possible without error correction. The method enables universal quantum simulation, including simulation of the time-dependent Schrodinger equation, and we argue that a 1000-qubit SES processor should be capable of achieving quantum speedup relative to a petaflop supercomputer. We speculate on the utility and practicality of such a simulator for atomic and molecular collision physics. [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