SEARCH · Search NASA

Results for “parallel cluster”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 433 records · Page 24

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Scale-up Unlearnable Examples Learning with High-performance Computing

Recent advancements in AI models, like ChatGPT, are structured to retain user interactions, which could inadvertently include sensitive healthcare data. In the healthcare field, particularly when radiologists use AI-driven diagnostic tools hosted on online platforms, there is a risk that medical imaging data may be repurposed for future AI training without explicit consent, spotlighting critical privacy and intellectual property concerns around healthcare data usage. Addressing these privacy challenges, a novel approach known as Unlearnable Examples (UEs) has been introduced, aiming to make data unlearnable to deep learning models. A prominent method within this area, called Unlearnable Clustering (UC), has shown improved UE performance with larger batch sizes but was previously limited by computational resources (e.g., a single workstation). To push the boundaries of UE performance with theoretically unlimited resources, we scaled up UC learning across various datasets using Distributed Data Parallel (DDP) training on the Summit supercomputer. Our goal was to examine UE efficacy at high-performance computing (HPC) levels to prevent unauthorized learning and enhance data security, particularly exploring the impact of batch size on UE’s unlearnability. Utilizing the robust computational capabilities of the Summit, extensive experiments were conducted on diverse datasets such as Pets, MedMNist, Flowers, and Flowers102. Our findings reveal that both overly large and overly small batch sizes can lead to performance instability and affect accuracy. However, the relationship between batch size and unlearnability varied across datasets, highlighting the necessity for tailored batch size strategies to achieve optimal data protection. The use of Summit’s high-performance GPUs, along with the efficiency of the DDP framework, facilitated rapid updates of model parameters and consistent training across nodes. Our results underscore the critical role of selecting appropriate batch sizes based on the specific characteristics of each dataset to prevent learning and ensure data security in deep learning applications. The source code is publicly available at https: // github. com/ hrlblab/ UE_ HPC .

Zhu, Yanfan [Vanderbilt University, Nashville, TN,↗

NASA NTRS2010

Whistler Waves Driven by Anisotropic Strahl Velocity Distributions: Cluster Observations

Observed properties of the strahl using high resolution 3D electron velocity distribution data obtained from the Cluster/PEACE experiment are used to investigate its linear stability. An automated method to isolate the strahl is used to allow its moments to be computed independent of the solar wind core+halo. Results show that the strahl can have a high temperature anisotropy (T(perpindicular)/T(parallell) approximately > 2). This anisotropy is shown to be an important free energy source for the excitation of high frequency whistler waves. The analysis suggests that the resultant whistler waves are strong enough to regulate the electron velocity distributions in the solar wind through pitch-angle scattering

Vinas, A.F.↗

DOE OSTI2024

Massively scalable workflows for quantum chemistry: BigChem and ChemCloud

Electronic structure theory, i.e., quantum chemistry, is the fundamental building block for many problems in computational chemistry. Here we present a new distributed computing framework (BigChem), which allows for an efficient solution of many quantum chemistry problems in parallel. BigChem is designed to be easily composable and leverages industry-standard middleware (e.g., Celery, RabbitMQ, and Redis) for distributed approaches to large scale problems. BigChem can harness any collection of worker nodes, including ones on cloud providers (such as AWS or Azure), local clusters, or supercomputer centers (and any mixture of these). BigChem builds upon MolSSI packages, such as QCEngine to standardize the operation of numerous computational chemistry programs, demonstrated here with Psi4, xtb, geomeTRIC, and TeraChem. BigChem delivers full utilization of compute resources at scale, offers a programable canvas for designing sophisticated quantum chemistry workflows, and is fault tolerant to node failures and network disruptions. We demonstrate linear scalability of BigChem running computational chemistry workloads on up to 125 GPUs. Finally, we present ChemCloud, a web API to BigChem and successor to TeraChem Cloud. ChemCloud delivers scalable and secure access to BigChem over the Internet.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2019

A cloud computing approach to massively-parallel distributed stochastic optimization with application to geomechanical reservoir characterization

Many search strategies used to optimize parameters during hydrologic model calibration are inherently limited in that simulations must be performed sequentially in order for the algorithm to converge. As a result, the overall time required to reach a solution increases linearly with the number of steps necessary. In contrast, sampling strategies that attempt to approximate the posterior distribution of the parameters do not require the model simulations to be run in sequence and therefore invite the use of a massively parallel approach across heterogeneous computing environments. We apply a cloud computing methodology to this problem where a head node running on Amazon Web Services continuously produces candidate models that are distributed to arbitrary compute environments to run the simulation, in our case Clemson’s Palmetto cluster computer. Thus the time required to obtain a sufficient number of samples to produce an adequate approximation of the posterior distribution of the parameters is limited by the number of available compute sites. Monte Carlo sampling of the posterior space would be duplicative and inefficient, thus we combine Latin hypercube methods with genetic algorithms to ensure that no large regions of the parameter space are left unexplored, and that high probability regions are investigated thoroughly. We independently assess the fit to different types of data sets without the need to construct a single objective function. Sets of Pareto optimal models are particularly valuable when the quality and information content of data is poorly known or conceptual model errors exist in our simulation. We illustrate our approach for a problem where in situ strain data are used to calibrate the parameters of an oil reservoir in the Avant oil field in Oklahoma. We use well injection and shut-in to stress a deep (500m) confined formation, and observe the geomechanical response using borehole strainmeters installed in the shallow subsurface (30m). These data were then used to calibrate the poroelastic properties of the reservoir and caprock along with the geometry of an elliptic channel associated with the formation. Using our approach we are able to identify issues with model error as well as substantial tradeoffs that occur within the parameters.

inversion, stochastic, geomechanics, poroelastic, ↗

NASA NTRS2008

Adsorption of Water on JSC-1A Lunar Simulant Samples

Remote sensing probes sent to the moon in the 1990s indicated that water may exist in areas such as the bottoms of deep, permanently shadowed craters at the lunar poles, buried under regolith. Water is of paramount importance for any lunar exploration and colonization project which would require self-sustainable systems. Therefore, investigating the interaction of water with lunar regolith is pertinent to future exploration. The lunar environment can be approximated in ultra-high vacuum systems such as those used in thermal desorption spectroscopy (TDS). Questions about water dissociation, surface wetting, degree of crystallization, details of water-ice transitions, and cluster formation kinetics can be addressed by TDS. Lunar regolith specimens collected during the Apollo missions are still available though precious, so testing with simulant is required before applying to use lunar regolith samples. Hence, we used for these studies JSC-1a, mostly an aluminosilicate glass and basaltic material containing substantial amounts of plagioclase, some olivine and traces of other minerals. Objectives of this project include: 1) Manufacturing samples using as little raw material as possible, allowing the use of surface chemistry and kinetics tools to determine the feasibility of parallel studies on regolith, and 2) Characterizing the adsorption kinetics of water on the regolith simulant. This has implications for the probability of finding water on the moon and, if present, for recovery techniques. For condensed water films, complex TDS data were obtained containing multiple features, which are related to subtle rearrangements of the water adlayer. Results from JSC-1a TDS studies indicate: 1) Water dissociation on JSC-1a at low exposures, with features detected at temperatures as high as 450 K and 2) The formation of 3D water clusters and a rather porous condensed water film. It appears plausible that the sub- m sized particles act as nucleation centers.

Goering, John↗

DOE OSTI2026

Spatially Accelerated Winding Numbers for Curved Geometry

The generalized winding number (GWN) is a scalar field that supports robust containment queries on curved geometry, including non-watertight, overlapping, and nested boundary representations. While queries can be easily parallelized over samples, direct evaluation on parametric curves and surfaces remains costly for large and complex models. Fast, state-of-the-art GWN approaches leverage a spatial index to approximate the GWN, typically coupled with a Taylor expansion which approximates the GWN contribution for far clusters of geometric primitives. However, such methods operate only on discrete inputs such as triangle meshes and point clouds, and would introduce containment errors near boundaries if applied to curved input. We extend support for fast GWN evaluation over arbitrary collections of NURBS curves in 2D and trimmed NURBS patches in 3D via a Bounding Volume Hierarchy that stores efficiently precomputed moment data in the hierarchy nodes. When querying the hierarchy, approximations for far clusters are used alongside direct evaluation for nearby NURBS primitives, achieving sub-linear complexity while preserving the geometric features in the vicinity of the query point. Central to our performance improvements is an adaptive subdivision strategy for NURBS primitives during a preprocessing phase, creating better spatial partitions while retaining the same accuracy for containment decisions as a direct evaluation. We demonstrate the performance and accuracy of our approach across a large collection of 2D and 3D datasets.

Computer science↗

DOE OSTI2019

Implementation of analytic gradients for CCSD and EOM-CCSD using Cholesky decomposition of the electron-repulsion integrals and their derivatives: Theory and benchmarks

We present a general formulation of analytic nuclear gradients for the coupled-cluster with single and double substitution (CCSD) and equation-of-motion (EOM) CCSD energies computed using Cholesky decomposition (CD) representations of the electron repulsion integrals. By rewriting the correlated energy and response equations such that the storage of the largest four-index intermediates is eliminated, CD leads to a significant reduction in disk storage requirements, reduced I/O penalties, and an improved parallel performance. CD thus extends the scope of the systems that can be treated by (EOM-)CCSD methods, although analytic gradients in the framework of CD are needed to extend the applicability of (EOM-)CCSD methods in the context of geometry optimizations. This paper presents a formulation of analytic (EOM-)CCSD gradient within the CD framework and reports on the salient details of the corresponding implementation. The accuracy and the capabilities of analytic CD-based (EOM-)CCSD gradients are illustrated by benchmark calculations and several illustrative examples.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2023

Molecular dynamics study of primary damage in the near-surface region in nickel

In this work, we carried out a large-scale molecular dynamics (MD) simulation to elucidate the effect of the free surface on the defect production of displacement cascades in pure nickel. These MD simulations were performed both in the bulk and near-surface regions with primary knock-on atom (PKA) energies of E PKA = 1, 5, or 10 keV and at temperatures T = 300, 425, or 525 K. Additionally, for every (T, E PKA ), near-surface cascade simulations were performed as a function of depth. In the near-surface simulations, assuming isotropic neutron fluence, the PKAs were initiated in random directions, including toward and parallel to the free surface. For both the near-surface and bulk cascades, the effect of E PKA and T on the defect production is similar. In both cases, the defect production increases with E PKA , but the temperature has minimal effect. However, the production and clustering of vacancies are higher for near-surface cascades, and they decrease with increasing depth. In contrast, the production and clustering of self-interstitial atoms are lower and increase with depth. Eventually, the production and clustering of both defect types approach bulk cascade-like behavior, and the depth at which this occurs increases with E PKA , but is independent of temperature.

36 MATERIALS SCIENCE↗

NASA NTRS1978

A simple method for obtaining a three-dimensional proton distribution function from Voyager plasma data

The main sensor of the Vogager plasma experiment consists of a cluster of three, modulated-grid Faraday cups whose normals are arranged symmetrically about the symmetry axis of the cluster at an angle of 20 degrees to that axis. In interplanetary space, each cup explores the positive ion distribution by accepting particles from contiguous slices in velocity space. The slices are narrow in the direction of the normal to the modulating grid but are broad in planes parallel to that grid. The resulting three sets of measurements can be combined to yield the three-dimensional distribution function in the following way: the distribution function is assumed to be gyrotropic. For each value of speed in a frame of reference moving with the bulk velocity of the solar wind, the variation of the distribution function with angle from the field direction is represented by a series of Legendre polynomials. Effects such as double-streaming and heat flow can be well represented by using only the first three terms of the series which are fully specified by the measurements. Examples of the use of this method in the analysis of Voyager data are shown.

Olbert, S.↗

NASA NTRS1988

Investigations on the usefulness of the Massively Parallel Processor for study of electronic properties of atomic and condensed matter systems

The usefulness of the Massively Parallel Processor (MPP) for investigation of electronic structures and hyperfine properties of atomic and condensed matter systems was explored. The major effort was directed towards the preparation of algorithms for parallelization of the computational procedure being used on serial computers for electronic structure calculations in condensed matter systems. Detailed descriptions of investigations and results are reported, including MPP adaptation of self-consistent charge extended Hueckel (SCCEH) procedure, MPP adaptation of the first-principles Hartree-Fock cluster procedure for electronic structures of large molecules and solid state systems, and MPP adaptation of the many-body procedure for atomic systems.

Das, T. P.↗

DOE OSTI2020

Supercomputer-Based Ensemble Docking Drug Discovery Pipeline with Application to Covid-19

In this work, we present a supercomputer-driven pipeline for in silico drug discovery using enhanced sampling molecular dynamics (MD) and ensemble docking. Ensemble docking makes use of MD results by docking compound databases into representative protein binding-site conformations, thus taking into account the dynamic properties of the binding sites. We also describe preliminary results obtained for 24 systems involving eight proteins of the proteome of SARS-CoV-2. The MD involves temperature replica exchange enhanced sampling, making use of massively parallel supercomputing to quickly sample the configurational space of protein drug targets. Using the Summit supercomputer at the Oak Ridge National Laboratory, more than 1 ms of enhanced sampling MD can be generated per day. We have ensemble docked repurposing databases to 10 configurations of each of the 24 SARS-CoV-2 systems using AutoDock Vina. Comparison to experiment demonstrates remarkably high hit rates for the top scoring tranches of compounds identified by our ensemble approach. We also demonstrate that, using Autodock-GPU on Summit, it is possible to perform exhaustive docking of one billion compounds in under 24 h. Finally, we discuss preliminary results and planned improvements to the pipeline, including the use of quantum mechanical (QM), machine learning, and artificial intelligence (AI) methods to cluster MD trajectories and rescore docking poses.

60 APPLIED LIFE SCIENCES↗

NASA NTRS2012

Observations of How Magnetofluid Turbulence Dissipates at Small Scales

The solar wind is a turbulent magneto fluid that can be studied intensively at multiple scales. Investigations using single spacecraft have revealed much about the properties of the solar wind throughout the heliosphere (from 0.3 AU to 100 AU). More recently, data from multiple spacecraft have provided further details of both the statistical properties of the turbulence and its small-scale structure. In particular, high time resolution magnetic field measurements from the four Cluster spacecrafl have led to the conclusion that at spatial scales of order the proton inertial length and smaller, the turbulence becomes strongly anisotropic and the power in fluctuations that are perpendicular to the (local) magnetic field is measured to be much larger than that in fluctuations that are parallel to the magnetic field. As the spatial scales approach the electron inertial length, the power is almost completely dissipated. Various analysis techniques and theoretical ideas have been put forward to account for the properties of those measurements. The talk will describe the current state of observations, theory and simulations.

Goldstein, M. L.↗

NASA NTRSDate not supplied

A Study of Parallel Scalability and Dynamic Workload Balancing in GlennICE

The Glenn Icing Computational Environment (GlennICE) is a computational tool designed to calculate ice growth on complex three-dimensional geometries. It utilizes user-supplied computational fluid dynamics solutions for the geometry of interest. Key developments include advancements in convergence of collection efficiency, trajectory optimization, and refinement methodology. These improvements have significantly enhanced GlennICE’s efficiency for practical engineering applications. A recent study focused on benchmarking GlennICE’s scalability in a parallel environment using static scheduling. Findings indicated a potential twofold increase in efficiency through workload balance enhancements. This paper presents an analysis of the solver’s new workload balancing improvements, incorporating shared memory and dynamic scheduling routines. Results demonstrate a highly efficient and consistent algorithm across high-performance computing clusters.

Computational Icing↗

NASA NTRSDate not supplied

A Study of Parallel Scalability and Dynamic Workload Balancing in GlennICE

Computational Icing↗

DOE OSTI2023

Structural evidence for intermediates during O 2 formation in photosystem II

In natural photosynthesis, the light-driven splitting of water into electrons, protons and molecular oxygen forms the first step of the solar-to-chemical energy conversion process. The reaction takes place in photosystem II, where the Mn 4 CaO 5 cluster first stores four oxidizing equivalents, the S 0 to S 4 intermediate states in the Kok cycle, sequentially generated by photochemical charge separations in the reaction center and then catalyzes the O–O bond formation chemistry. Here, we report room temperature snapshots by serial femtosecond X-ray crystallography to provide structural insights into the final reaction step of Kok’s photosynthetic water oxidation cycle, the S 3 →[S 4 ]→S 0 transition where O 2 is formed and Kok’s water oxidation clock is reset. Our data reveal a complex sequence of events, which occur over micro- to milliseconds, comprising changes at the Mn 4 CaO 5 cluster, its ligands and water pathways as well as controlled proton release through the hydrogen-bonding network of the Cl1 channel. Importantly, the extra O atom O x , which was introduced as a bridging ligand between Ca and Mn1 during the S 2 →S 3 transition disappears or relocates in parallel with Y z reduction starting at approximately 700 μs after the third flash. The onset of O 2 evolution, as indicated by the shortening of the Mn1–Mn4 distance, occurs at around 1,200 μs, signifying the presence of a reduced intermediate, possibly a bound peroxide.

59 BASIC BIOLOGICAL SCIENCES↗

NASA NTRS2005

In Vitro Experimental Model to Investigate the Biological Effects across the Bragg Curve of High-LET Radiation

The space environment consists of a varying field of radiation particles including high energy ions, with a spacecrafts shielding material providing the only major protection to astronauts from harmful exposure. Unlike lowLET gamma or Xrays, the presence of shielding does not always reduce the radiation risks for energetic charged particle exposure since the dose delivered by the charged particle increases sharply as the particle approaches the end of its range, a position known as the Bragg peak and the correlating spatial dose distribution identified as the Bragg curve. The Bragg curve does not necessarily represent the biological damage along the particle traversal since biological effects are influenced by the track structure of both primary and secondary particles. Therefore, the biological Bragg curve is dependent of the energy and the type of the primary particle, and may vary for different biological endpoints. Here we describe a unique irradiation geometry and experimental system to measure the biological response across the Bragg curve in one consistent biological sample. Polyethylene shielding was used to achieve a Bragg curve distribution with the beam geometry parallel to a monolayer of fibroblast cells. We present data that highlights the differential formation of DNA double strand breaks (DSBs) and chromosomal deletions across the Bragg curve in human fibroblasts irradiated with 600 MeV/nucleon iron ion beams. Qualitative analyses of gammaH2AX fluorescence, a known marker of DSBs, indicated potentially increased clustering of DNA damage before the Bragg peak, enhanced homogenous distribution at the peak, and provided visual evidence of high linear energy transfer (LET) particle traversal of cells beyond the Bragg peak in agreement with one-dimensional transport approximations. A biological response curve generated for micronuclei induction across the Bragg curve for 600 MeV/n Fe ions did not reveal an increase in the yield of micronuclei at the Bragg peak location. Assessment of such biological parameters employing the described in vitro experimental system may provide improved platforms to measure a number of biological consequences of shielding materials across the Bragg curve for high charge and energy (HZE) ions.

Desai, N.↗

DOE OSTI2021

A Case Study of LLVM-Based Analysis for Optimizing SIMD Code Generation

This paper presents a methodology for using LLVM-based tools to tune the DCA++ (dynamical cluster approximation) application that targets the new ARM A64FX processor. The goal is to describe the changes required for the new architecture and generate efficient single instruction/multiple data (SIMD) instructions that target the new Scalable Vector Extension instruction set. During manual tuning, the authors used the LLVM tools to improve code parallelization by using OpenMP SIMD, refactored the code and applied transformation that enabled SIMD optimizations, and ensured that the correct libraries were used to achieve optimal performance. By applying these code changes, code speed was increased by 1.98× and 78 GFlops were achieved on the A64FX processor. The authors aim to automatize parts of the efforts in the OpenMP Advisor tool, which is built on top of existing and newly introduced LLVM tooling.

Huber, Joseph↗

DOE OSTI2024

Direct conversion of methane to aromatics and hydrogen via a heterogeneous trimetallic synergistic catalyst

Abstract Non-oxidative methane dehydro-aromatization reaction can co-produce hydrogen and benzene effectively on a molybdenum-zeolite based thermochemical catalyst, which is a very promising approach for natural-gas upgrading. However, the low methane conversion and aromatics selectivity and weak durability restrain the realistic application for industry. Here, a mechanism for enhancing catalysis activity on methane activation and carbon-carbon bond coupling has been found to promote conversion and selectivity simultaneously by adding platinum–bismuth alloy cluster to form a trimetallic catalyst on zeolite (Pt-Bi/Mo/ZSM-5). This bimetallic alloy cluster has synergistic interaction with molybdenum: the formed CH 3 * from Mo 2 C on the external surface of zeolite can efficiently move on for C-C coupling on the surface of Pt-Bi particle to produce C 2 compounds, which are the key intermediates of oligomerization. This pathway is parallel with the catalysis on Mo inside the cage. This catalyst demonstrated 18.7% methane conversion and 69.4% benzene selectivity at 710 °C. With 95% methane/5% nitrogen feedstock, it exhibited robust stability with slow deactivation rate of 9.3% after 2 h and instant recovery of 98.6% activity after regeneration in hydrogen. The enhanced catalytic activity is strongly associated with synergistic interaction with Mo and ligand effects of alloys by extensive mechanism studies and DFT calculation.

03 NATURAL GAS↗