A5063517452- Lattice Boltzmann Simulation Studies
- Distributed and Parallel Computing Systems
- Parallel Computing and Optimization Techniques
- Advanced Data Storage Technologies
- Advanced Chemical Physics Studies
- Spectroscopy and Quantum Chemical Studies
- Blood properties and coagulation
- Cardiovascular Health and Disease Prevention
- Quantum, superfluid, helium dynamics
- Generative Adversarial Networks and Image Synthesis
- ECG Monitoring and Analysis
- Cardiovascular Function and Risk Factors
- Cardiac electrophysiology and arrhythmias
- Advanced Battery Materials and Technologies
- Advancements in Battery Materials
- Spacecraft and Cryogenic Technologies
- Ionic liquids properties and applications
- Scientific Computing and Data Management
- Cloud Computing and Resource Management
- Nuclear reactor physics and engineering
- Quantum Computing Algorithms and Architecture
- Nanowire Synthesis and Applications
- Numerical methods for differential equations
- Silicon Nanostructures and Photoluminescence
- Advanced Battery Technologies Research
Lawrence Livermore National Laboratory
2015-2024
Michigan State University
2021
Bioscience (China)
2006
A series of Car-Parrinello (CP) molecular dynamics simulations water are presented, aimed at assessing the accuracy density functional theory in describing structural and dynamical properties ambient conditions. We found negligible differences obtained using Perdew-Burke-Ernzerhof or Becke-Lee-Yang-Parr exchange correlation energy functionals; we also that size effects, although not fully when 32 molecule cells, rather small. In addition, identified a wide range values fictitious electronic...
Lithium-ion battery performance is strongly influenced by the ionic conductivity of electrolyte, which depends on speed at Li ions migrate across cell and relates to their solvation structure. The choice solvent can greatly impact both diffusivity ions. We use first principles molecular dynamics examine diffusion in bulk organic solvents ethylene carbonate (EC), ethyl methyl (EMC), a mixture EC/EMC. find that are solvated either carbonyl or ether oxygen atoms sometimes PF$_6^-$ anion. Li$^+$...
As noted in Wikipedia, skin the game refers to having ‘incurred risk by being involved achieving a goal’, where ‘ is synecdoche for person involved, and metaphor actions on field of play under discussion’. For exascale applications development US Department Energy Exascale Computing Project, nothing could be more apt, with delivering comprehensive science-based computational that effectively exploit high-performance computing technologies provide breakthrough modelling simulation data...
Explicit integrators for real-time propagation of time-dependent Kohn-Sham equations are compared regarding their suitability performing large-scale simulations. Four algorithms implemented and assessed both stability accuracy within a plane-wave pseudopotential framework, employing the adiabatic approximation to exchange-correlation functional. Simulation results single sodium atom embedded in bulk magnesium oxide discussed. While first-order Euler scheme second-order finite-difference...
The potential impact of blood flow simulations on the diagnosis and treatment patients suffering from vascular disease is tremendous. Empowering models full arterial tree can provide insight into diseases such as hypertension enables study influence local factors global hemodynamics. We present a new, highly scalable implementation lattice Boltzmann method which addresses key challenges multiscale coupling, limited memory capacity bandwidth, robust load balancing in complex geometries....
Path-integral Monte Carlo calculations of the superfluid density throughout 4He droplets doped with linear impurities are presented. After deriving a local estimator for distribution, we find decreased response in cylindrically symmetric region first solvation layer. The helium this has transition temperature similar to that two-dimensional system and may be responsible previously unexplained experimental Q-branch measurements.
First-principles simulations of high-Z metallic systems using the Qbox code on BlueGene/L supercomputer demonstrate unprecedented performance and scaling for a quantum simulation code. Specifically designed to take advantage massively-parallel like BlueGene/L, demonstrates excellent parallel efficiency peak performance. A sustained 207.3 TFlop/s was measured 65,536 nodes, corresponding 56.5% theoretical full machine all 128k CPUs.
Abstract Comorbidities such as anemia or hypertension and physiological factors related to exertion can influence a patient’s hemodynamics increase the severity of many cardiovascular diseases. Observing quantifying associations between these be difficult due multitude co-existing conditions blood flow parameters in real patient data. Machine learning-driven, physics-based simulations provide means understand how potentially correlated may affect particular patient. Here, we use combination...
The placement of tasks in a parallel application on specific nodes supercomputer can significantly impact performance. Traditionally, this task mapping has focused reducing the distance between communicating physical network. This minimizes number hops that point-to-point messages travel and thus reduces link sharing contention. However, for applications use collectives over sub-communicators, heuristic may not be optimal. Many benefit from an increase bandwidth even at cost hop count,...
AbstractWe have developed the capability to rapidly simulate cardiac electrophysiological phenomena in a human heart discretised at resolution comparable with length of myocyte. Previous scientific investigation has generally invoked simplified geometries or coarse-resolution hearts, simulation duration limited 10s heartbeats. Using state-of-the-art high-performance computing techniques coupled one most powerful computers available (the 20 PFlop/s IBM BlueGene/Q Lawrence Livermore National...
We demonstrate that the Qbox code supports unprecedented large-scale First-Principles Molecular Dynamics (FPMD) applications on BlueGene/L supercomputer. is an FPMD implementation specifically designed for parallel platforms such as BlueGene/L. Strong scaling tests a Materials Science application show 86% efficiency between 1024 and 32,768 CPUs. Measurements of performance by means hardware counters 36% peak FPU can be attained.
The placement of tasks in a parallel application on specific nodes supercomputer can significantly impact performance. Traditionally, this task mapping has focused reducing the distance between communicating physical network. This minimizes number hops that point-to-point messages travel and thus reduces link sharing contention. However, for applications use collectives over sub-communicators, heuristic may not be optimal. Many benefit from an increase bandwidth even at cost hop count,...
Simulations of the passage eukaryotic cells through a constricted channel aid in studying properties cancer and their transport bloodstream. Compound capsules, which explicitly model outer cell membrane nuclear lamina, have potential to improve computational fidelity. However, general simulations compound capsules transiting microchannel not been conducted influence capsule on performance is well known. In this study, we extend parallel hemodynamics application simulate fluid-structure...
The US Department of Energy Office Science and the National Nuclear Security Administration initiated Exascale Computing Project (ECP) in 2016 to prepare mission-relevant applications scientific software for delivery exascale computers starting 2023. ECP currently supports 24 efforts directed at specific six supporting co-design projects. These application projects contain 62 codes that are implemented three high-level languages—C, C++, Fortran—and use 22 combinations graphical processing...
First-principles molecular dynamics and quantum Monte Carlo techniques are employed to gain insight into the effect of preparation conditions on structural optical properties silicon nanoparticles. Our results demonstrate that (i) kinetically limited nanostructures form different core structures than bulk-derived crystalline clusters, (ii) type structure forms depends how cluster is passivated during synthesis, (iii) good agreement with measured gaps can be obtained for nanoparticles from...
The effect of preparation conditions on the structural and optical properties silicon nanoparticles is investigated. Nanoscale reconstructions, unique to curved nanosurfaces, are presented for nanocrystals shown have lower energy larger gaps than bulk-derived structures. We find that high-temperature synthesis processes can produce metastable noncrystalline nanostructures with different core structures crystalline clusters. type structure forms from a given process may depend passivation...
With the current rate of progress in quantum computing technologies, systems with more than 50 qubits will soon become reality. Computing ideal state amplitudes for circuits such and larger sizes is a fundamental step to assess both correctness, performance, scaling behavior algorithms fidelities devices. However, resource requirements calculations on classical computers grow exponentially. We show that deferring tensor contractions can extend boundaries what be computed systems. To...
Tracking hemodynamic responses to treatment and stimuli over long periods remains a grand challenge. Moving from established single-heartbeat technology longitudinal profiles would require continuous data describing how the patient's state evolves, new methods extend temporal domain which flow is sampled, high-throughput computing resources. While personalized digital twins can accurately measure 3D hemodynamics several heartbeats, state-of-the-art hundreds of years wallclock time on...
First-Principles Molecular Dynamics (FPMD) methods, although powerful, are notoriously expensive computationally due to the quantum modeling of electrons. Traditional FPMD approaches have typically been limited a few thousand atoms at most, O(N3) or worse solver complexity and large amount communication required for highly parallel implementations. Attempts lower often introduced uncontrolled approximations systematic errors. Using robust new algorithm, we developed an O(N) electronic...