Publications

Davis, Jean-Paul D.; Lane, James M.; Thompson, Aidan P.; Drake, Richard R.; Weirs, Vincent G.; Flicker, Dawn G.; Robbins, Joshua R.; Lemke, Raymond W.; Martin, Matthew; Alexander, Charles S.; Haill, Thomas A.; Ao, Tommy A.; Dolan, Daniel H.

Abstract not provided.

Thompson, Aidan P.; Trott, Christian R.

Within the EXAALT project, the SNAP [1] approach is being used to develop high accuracy potentials for use in large-scale long-time molecular dynamics simulations of materials behavior. In particular, we have developed a new SNAP potential that is suitable for describing the interplay between helium atoms and vacancies in high-temperature tungsten[2]. This model is now being used to study plasma-surface interactions in nuclear fusion reactors for energy production. The high-accuracy of SNAP potentials comes at the price of increased computational cost per atom and increased computational complexity. The increased cost is mitigated by improvements in strong scaling that can be achieved using advanced algorithms [3].

Crozier, Paul C.; Plimpton, Steven J.; Thompson, Aidan P.

Abstract not provided.

Lo, Chi S.; Thompson, Aidan P.

Abstract not provided.

Foiles, Stephen M.; Thompson, Aidan P.; Swiler, Laura P.; Trott, Christian R.

Abstract not provided.

Macromolecules

Budzien, Joanne L.; Lo, Chi S.; Thompson, Aidan P.

Abstract not provided.

Wood, Mitchell A.; Plimpton, Steven J.; Thompson, Aidan P.; Perez, Danny P.; Niklasson, Anders N.

Abstract not provided.

Thompson, Aidan P.

Abstract not provided.

International Journal of High Performance Computing Applications

Levy, Scott; Ferreira, Kurt; Bridges, Patrick G.; Thompson, Aidan P.; Trott, Christian R.

Building the next-generation of extreme-scale distributed systems will require overcoming several challenges related to system resilience. As the number of processors in these systems grow, the failure rate increases proportionally. One of the most common sources of failure in large-scale systems is memory. In this paper, we propose a novel runtime for transparently exploiting memory content similarity to improve system resilience by reducing the rate at which memory errors lead to node failure. We evaluate the viability of this approach by examining memory snapshots collected from eight high-performance computing (HPC) applications and two important HPC operating systems. Based on the characteristics of the similarity uncovered, we conclude that our proposed approach shows promise for addressing system resilience in large-scale systems.

Deng, Jie D.; Erickson, Lindsay C.; Plimpton, Steven J.; Thompson, Aidan P.; Zhou, Xiaowang Z.; Zimmerman, Jonathan A.

Many of the most important and hardest-to-solve problems related to the synthesis, performance, and aging of materials involve diffusion through the material or along surfaces and interfaces. These diffusion processes are driven by motions at the atomic scale, but traditional atomistic simulation methods such as molecular dynamics are limited to very short timescales on the order of the atomic vibration period (less than a picosecond), while macroscale diffusion takes place over timescales many orders of magnitude larger. We have completed an LDRD project with the goal of developing and implementing new simulation tools to overcome this timescale problem. In particular, we have focused on two main classes of methods: accelerated molecular dynamics methods that seek to extend the timescale attainable in atomistic simulations, and so-called 'equation-free' methods that combine a fine scale atomistic description of a system with a slower, coarse scale description in order to project the system forward over long times.

Physical Review B

Ellis, J.A.; Fiedler, L.; Popoola, G.A.; Modine, N.A.; Stephens, John A.; Thompson, Aidan P.; Cangi, A.; Rajamanickam, Sivasankaran R.

We present a numerical modeling workflow based on machine learning which reproduces the total energies produced by Kohn-Sham density functional theory (DFT) at finite electronic temperature to within chemical accuracy at negligible computational cost. Based on deep neural networks, our workflow yields the local density of states (LDOS) for a given atomic configuration. From the LDOS, spatially resolved, energy-resolved, and integrated quantities can be calculated, including the DFT total free energy, which serves as the Born-Oppenheimer potential energy surface for the atoms. We demonstrate the efficacy of this approach for both solid and liquid metals and compare results between independent and unified machine-learning models for solid and liquid aluminum. Our machine-learning density functional theory framework opens up the path towards multiscale materials modeling for matter under ambient and extreme conditions at a computational scale and cost that is unattainable with current algorithms.

Ellis, J.A.; Rajamanickam, Sivasankaran R.; Modine, N.A.; Thompson, Aidan P.; Stephens, John A.; Cangi, Attila -.

Abstract not provided.

Modine, N.A.; Stephens, John A.; Swiler, Laura P.; Thompson, Aidan P.; Vogel, Dayton J.; Cangi, Attila C.; Feilder, Lenz F.; Rajamanickam, Sivasankaran R.

The focus of this project is to accelerate and transform the workflow of multiscale materials modeling by developing an integrated toolchain seamlessly combining DFT, SNAP, LAMMPS, (shown in Figure 1-1) and a machine-learning (ML) model that will more efficiently extract information from a smaller set of first-principles calculations. Our ML model enables us to accelerate first-principles data generation by interpolating existing high fidelity data, and extend the simulation scale by extrapolating high fidelity data (10² atoms) to the mesoscale (10⁴ atoms). It encodes the underlying physics of atomic interactions on the microscopic scale by adapting a variety of ML techniques such as deep neural networks (DNNs), and graph neural networks (GNNs). We developed a new surrogate model for density functional theory using deep neural networks. The developed ML surrogate is demonstrated in a workflow to generate accurate band energies, total energies, and density of the 298K and 933K Aluminum systems. Furthermore, the models can be used to predict the quantities of interest for systems with more number of atoms than the training data set. We have demonstrated that the ML model can be used to compute the quantities of interest for systems with 100,000 Al atoms. When compared with 2000 Al system the new surrogate model is as accurate as DFT, but three orders of magnitude faster. We also explored optimal experimental design techniques to choose the training data and novel Graph Neural Networks to train on smaller data sets. These are promising methods that need to be explored in the future.

Thompson, Aidan P.; Slepoy, Alexander S.; Davidson, George S.; Peters, Michael D.

Abstract not provided.

Wills, Ann E.; Thompson, Aidan P.; Raman, Sumathy R.

We establish an atomistic view of the high- and low-temperature phases of iron/steel as well as some elements of the phase transition between these phases on cooling. In particular we examine the 4 most common orientation relationships between the high temperature austen- ite and low-temperature ferrite phases seen in experiment. With a thorough understanding of these relationships we are prepared to set up various atomistic simulations, using tech- niques such as Density Functional Theory and Molecular Dynamics, to further study the phase transition, in particular, quantities needed for Phase Field Modeling, such as the free energies of bulk phases and the phase transition front propagation velocity.

Ferreira, Kurt; Thompson, Aidan P.; Trott, Christian R.; Levy, Scott L.

Abstract not provided.

Budzien, Joanne L.; Lo, Chi S.; Grest, Gary S.; Thompson, Aidan P.

Abstract not provided.

Shan, Tzu-Ray S.; Thompson, Aidan P.; Yarrington, Cole Y.; Kittell, David E.; Wixom, Ryan R.

Abstract not provided.

Shan, Tzu-Ray S.; Thompson, Aidan P.; Wills, Ann E.

Abstract not provided.

Thompson, Aidan P.; Shan, Tzu-Ray S.; Wixom, Ryan R.; Yarrington, Cole Y.

Abstract not provided.

Thompson, Aidan P.

Abstract not provided.

Thompson, Aidan P.; Schultz, Peter A.; Crozier, Paul C.; Moore, Stan G.; Swiler, Laura P.; Stephens, John A.; Trott, Christian R.; Foiles, Stephen M.; Tucker, Garritt J.

This report summarizes the result of LDRD project 12-0395, titled "Automated Algorithms for Quantum-level Accuracy in Atomistic Simulations." During the course of this LDRD, we have developed an interatomic potential for solids and liquids called Spectral Neighbor Analysis Poten- tial (SNAP). The SNAP potential has a very general form and uses machine-learning techniques to reproduce the energies, forces, and stress tensors of a large set of small configurations of atoms, which are obtained using high-accuracy quantum electronic structure (QM) calculations. The local environment of each atom is characterized by a set of bispectrum components of the local neighbor density projected on to a basis of hyperspherical harmonics in four dimensions. The SNAP coef- ficients are determined using weighted least-squares linear regression against the full QM training set. This allows the SNAP potential to be fit in a robust, automated manner to large QM data sets using many bispectrum components. The calculation of the bispectrum components and the SNAP potential are implemented in the LAMMPS parallel molecular dynamics code. Global optimization methods in the DAKOTA software package are used to seek out good choices of hyperparameters that define the overall structure of the SNAP potential. FitSnap.py, a Python-based software pack- age interfacing to both LAMMPS and DAKOTA is used to formulate the linear regression problem, solve it, and analyze the accuracy of the resultant SNAP potential. We describe a SNAP potential for tantalum that accurately reproduces a variety of solid and liquid properties. Most significantly, in contrast to existing tantalum potentials, SNAP correctly predicts the Peierls barrier for screw dislocation motion. We also present results from SNAP potentials generated for indium phosphide (InP) and silica (SiO 2 ). We describe efficient algorithms for calculating SNAP forces and energies in molecular dynamics simulations using massively parallel computers and advanced processor ar- chitectures. Finally, we briefly describe the MSM method for efficient calculation of electrostatic interactions on massively parallel computers.

Thompson, Aidan P.; Schultz, Peter A.; Swiler, Laura P.; Trott, Christian R.; Tucker, Garritt T.; Foiles, Stephen M.

Abstract not provided.

Alexander, Charles S.; Thompson, Aidan P.; Chhabildas, Lalit C.; Furnish, Michael D.; Dolan, Daniel H.; Davis, Jean-Paul D.; Lemke, Raymond W.; Desjarlais, Michael P.; Hertel, Eugene S.

Abstract not provided.

Nuclear Fusion

Cusentino, Mary A.; Wood, M.A.; Thompson, Aidan P.

Erosion of the beryllium first wall material in tokamak reactors has been shown to result in transport and deposition on the tungsten divertor. Experimental studies of beryllium implantation in tungsten indicate that mixed W-Be intermetallic deposits can form, which have lower melting temperatures than tungsten and can trap tritium at higher rates. To better understand the formation and growth rate of these intermetallics, cumulative molecular dynamics (MD) simulations of both high and low energy beryllium deposition in tungsten were performed. In both cases, a W-Be mixed material layer (MML) emerged at the surface within several nanoseconds, either through energetic implantation or a thermally-activated exchange mechanism, respectively. While some ordering of the material into intermetallics occurred, fully ordered structures did not emerge from the deposition simulations. Targeted MD simulations of the MML to further study the rate of Be diffusion and intermetallic growth rates indicate that for both cases, the gradual re-structuring of the material into an ordered intermetallic layer is beyond accessible MD time scales(≼1 μs). However, the rapid formation of the MML within nanoseconds indicates that beryllium deposition can influence other plasma species interactions at the surface and begin to alter the tungsten material properties. Therefore, beryllium deposition on the divertor surface, even in small amounts, is likely to cause significant changes in plasma-surface interactions and will need to be considered in future studies.