Search NASA⌕ Search

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.

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

319 records · Page 18
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

iez: Resource Contention Aware Load Balancing for Large-Scale Parallel File Systems

Parallel I/O performance is crucial to sustaining scientific applications on large-scale High-Performance Computing (HPC) systems. However, I/O load imbalance in the underlying distributed and shared storage systems can significantly reduce overall application performance. There are two conflicting challenges to mitigate this load imbalance: (i) optimizing systemwide data placement to maximize the bandwidth advantages of distributed storage servers, i.e., allocating I/O resources efficiently across applications and job runs; and (ii) optimizing client-centric data movement to minimize I/O load request latency between clients and servers, i.e., allocating I/O resources efficiently in service to a single application and job run. Moreover, existing approaches that require application changes limit wide-spread adoption in commercial or proprietary deployments. We propose iez, an “end-to-end control plane” where clients transparently and adaptively write to a set of selected I/O servers to achieve balanced data placement. Our control plane leverages realtime load information for distributed storage server global data placement while our design model leverages trace-based optimization techniques to minimize I/O load request latency between clients and servers. We evaluate our proposed system on an experimental cluster for two common use cases: synthetic I/O benchmark IOR for large sequential writes and a scientific application I/O kernel, HACC-I/O. Results show read and write performance improvements of up to 34% and 32%, respectively, compared to the state of the art.

Wadhwa, Bharti↗
DOE OSTI2025

Discovery of a Ferromagnetic Nickel Chalcogenide Nanocluster Ni 3 S 3 H(PEt 3 ) 5

Atomically precise ligated nanoclusters (NC) are promising cluster-based materials with novel molecular architectures and tunable magnetic properties. Herein, the synthesis and characterization of a nickel sulfide NC Ni 3 S 3 H(PEt 3 ) 5 (PEt 3 = triethylphosphine) with distinct magnetic properties are reported. Magnetization measurements reveal its magnetic moment of 1.5 µ B in the solid phase, consistent with the existence of one unpaired electron predicted by density functional theory (DFT) calculations. Additionally, experimental measurements indicate the presence of ferromagnetic ordering within each Ni 3 S 3 H(PEt 3 ) 5 NC and strong coercivity at temperatures below 20 K. Ion mobility-mass spectrometry is employed in conjunction with DFT calculations and collision cross-section simulations to investigate the structure of the isolated Ni 3 S 3 H(PEt 3 ) 5 . Theoretical studies show that [Ni 3 S 3 H(PEt 3 ) 5 ] + has a planar Ni 3 S 3 core where three Ni atoms are arranged in a triangle with three bridging S atoms residing in the same plane. This structure is preserved in both solution and solid phases, which is confirmed by spectroscopic studies of Ni 3 S 3 H(PEt 3 ) 5 . Additionally, DFT calculations indicate that all spins at the Ni sites are aligned parallel, confirming the presence of ferromagnetic coupling. Overall, this study provides key insights into the structure and magnetic properties of Ni 3 S 3 H(PEt 3 ) 5 , which will facilitate the design of new NC-based magnetic materials.

Nickel sulfide nanocluster↗
DOE OSTI2020

Bridging the Transuranics with Uranium(IV) Sulfate Aqueous Species and Solid Phases

Isolating isomorphic compounds of tetravalent actinides (i.e., Th IV , U IV , Np IV , and Pu IV ) improve our understanding of the bonding behavior across the series, in addition to their relationship with tetravalent transition metals (Zr and Hf) and lanthanides (Ce). Similarities between these tetravalent metals are particularly illuminated in their hydrolysis and condensation behavior in aqueous systems, leading to polynuclear clusters typified by the hexamer [M IV 6 O 4 (OH) 4 ] 12+ building block. Prior studies have shown the predominance and coexistence of smaller species for Th IV (monomers, dimers, and hexamers) and larger species for U IV , Np IV , and Pu IV (including 38-mers and 70-mers). We show here that aqueous uranium(IV) sulfate also displays behavior similar to that of Th IV (and Zr IV ) in its isolated solidphase and solution speciation. Two single-crystal X-ray structures are described: a dihydroxide-bridged dimer (U 2 ) formulated as U 2 (OH) 2 (SO 4 ) 3 (H 2 O) 4 and a monomer-linked hexamer framework (U-U 6 ) as (U(H 2 O) 3.5 ) 2 U 6 O 4 (OH) 4 (SO 4 ) 10 (H 2 O) 9 . These structures are similar to those previously described for ThIV. Moreover, cocrystallization of monomer and dimer and of dimer and monomer–hexamer phases for both Th IV (prior) and U IV (current) indicates the coexistence of these species in solution. Because it was not possible to effectively study the sulfate-rich solutions via X-ray scattering from which U 2 and U-U 6 crystallized, we provide a parallel solution speciation study in low sulfate conditions, as a function of the pH. Raman spectroscopy, UV–vis spectroscopy, and small-angle X-ray scattering of these show decreasing sulfate binding, increased hydrolysis, increased species size, and increased complexity, with increasing pH. Furthermore, this study describes a bridge across the first half the actinide series, highlighting U IV similarities to Th IV , in addition to the previously known similarities to the transuranic elements.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

MemXCT: Design, Optimization, Scaling, and Reproducibility of X-Ray Tomography Imaging

Here, this work extends our previous research entitled "MemXCT: Memory-centric X-ray CT Reconstruction with Massive Parallelization" that was originally published at SC19 conference (Hidayetoglu et al., 2019) with reproducibility of the computational imaging performance. X-ray computed tomography (XCT) is regularly used at synchrotron light sources to study the internal morphology of materials at high resolution. However, experimental constraints, such as radiation sensitivity, can result in noisy or undersampled measurements. Further, depending on the resolution, sample size and data acquisition rates, the resulting noisy dataset can be in the order of terabytes. Advanced iterative reconstruction techniques can produce high-quality images from noisy measurements, but their computational requirements have made their use an exception rather than the rule. We propose a novel memory-centric approach that avoids redundant computations at the expense of additional memory complexity. We develop a memory-centric iterative reconstruction system, MemXCT, that uses an optimized SpMV implementation with two-level pseudo-Hilbert ordering and multi-stage input buffering. We evaluate MemXCT on various supercomputer architectures involving KNL and GPU. MemXCT can reconstruct a large (11Kx11K) mouse brain tomogram in 10 seconds using 4096 KNL nodes (256K cores). The results presented in our original article at the SC19 were based on large-scale supercomputing resources. The MemXCT application was selected for the Student Cluster Competition (SCC) Reproducibility Challenge and evaluated on a variety of cloud computing resources by universities around the world in the SC20 conference. We summarize the results of the top-ranked SCC Reproducibility Challenge teams and identify the most pertinent measures for ensuring the reproducibility of our experiments in this article.

47 OTHER INSTRUMENTATION↗
DOE OSTI2022

Galaxy Alignments with Surrounding Structure in the Sloan Digital Sky Survey

Using data from the Sloan Digital Sky Survey Legacy Survey, we study the alignment of luminous galaxies with spectroscopic data with the surrounding larger-scale structure as defined by galaxies with only photometric data. We find that galaxies from the red sequence have a statistically significant tendency for their apparent long axes to align parallel to the projected surrounding structure. Red galaxies more luminous than the median of our sample (M r < –21.78) have a mean alignment angle $\langle$Φ$\rangle$ < 45°, indicating preferred parallel alignment, at a significance level >4.5σ on projected scales 0.1 Mpc < r p ≤ 7.5 Mpc. Fainter red galaxies have $\langle$Φ$\rangle$ < 45° at a significance level >4.3σ at scales 1 Mpc < r p < 3 Mpc. At a projected scale r p = 3.0 Mpc, the mean alignment angle decreases steadily with increasing luminosity for red galaxies with M r ≲ –22.5, reaching $\langle$Φ$\rangle$ = 40$^{°}_{.}$49 ± 0$^{°}_{.}$56 for the most luminous 1% (M r ~ –23.57). Galaxies from the blue sequence show no statistically significant tendency for their axes to align with larger-scale structure, regardless of galaxy luminosity. Galaxies in higher-density regions do not show a statistically significant difference in the mean alignment angle from galaxies in lower-density regions; this holds true for the faint blue, luminous blue, faint red, and luminous red subsets.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Life cycle of streaks in the buffer layer of wall-bounded turbulence

In this work, streaks in the buffer layer of wall-bounded turbulence are tracked in time to study their life cycle. Spatially and temporally resolved direct numerical simulation data are used to analyze the strong wall-parallel movements conditioned to low-speed streamwise flow. The analysis of the streaks shows that there is a clear distinction between wall-attached and detached streaks, and that the wall-attached streaks can be further categorized into streaks that are contained in the buffer layer and the ones that reach the outer region. The results reveal that streaks are born in the buffer layer, coalescing with each other to create larger streaks that are still attached to the wall. Once the streak becomes large enough, it starts to meander due to the large streamwise-to-wall-normal aspect ratio, and consequently the elongation in the streamwise direction, which makes it more difficult for the streak to be oriented strictly in the streamwise direction. While the continuous interaction of the streaks allows the superstructure to span extremely long temporal and length scales, individual streak components are relatively small and short-lived. Tall-attached streaks eventually split into wall-attached and wall-detached components. These wall-detached streaks have a strong wall-normal velocity away from the wall, similar to ejections or bursts observed in the literature. Conditionally averaging the flow fields to these split events show that the detached streak has not only a larger wall-normal velocity compared to the wall-attached counterpart, it also has a larger (less negative) streamwise velocity, similar to the velocity field at the tip of a vortex cluster.

42 ENGINEERING↗
DOE OSTI2022

Machine learning–driven multiscale modeling reveals lipid-dependent dynamics of RAS signaling proteins

Significance Here we present an unprecedented multiscale simulation platform that enables modeling, hypothesis generation, and discovery across biologically relevant length and time scales to predict mechanisms that can be tested experimentally. We demonstrate that our predictive simulation-experimental validation loop generates accurate insights into RAS-membrane biology. Evaluating over 100,000 correlated simulations, we show that RAS–lipid interactions are dynamic and evolving, resulting in: 1) a reordering and selection of lipid domains in realistic eight-lipid bilayers, 2) clustering of RAS into multimers correlating with specific lipid fingerprints, 3) changes in the orientation of the RAS G-domain impacting its ability to interact with effectors, and 4) demonstration that RAS–RAS G-domain interfaces are nonspecific in these putative signaling domains.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2021

Equatorial Electronic Structure in the Uranyl Ion: Cs 2 UO 2 Cl 4 and Cs 2 UO 2 Br 4

Electric field gradient (EFG) tensors in the equatorial plane of the linear UO 2 2+ ion have been measured by nuclear magnetic resonance (NMR) and nuclear quadrupole resonance (NQR) experiments and computed by relativistic Kohn–Sham methods with and without environment embedding for Cs 2 UO 2 Cl 4 and Cs 2 UO 2 Br 4 . This approach expands the possibilities for probing the electronic structure in uranyl complexes beyond the strongly covalent U–O bonds. The combined analyses find that one of the two largest principal EFG tensor components at the halogen sites points along the U–X bond (X = Cl, Br), and the second is parallel to the UO 2 2+ ion; in Cs 2 UO 2 Cl 4 , the components are nearly equal in magnitude, whereas in Cs 2 UO 2 Br 4 , due to short-range bromide–cesium interactions, the equatorial component is dominant for one pair of Br sites and the axial component is larger for the second pair. Additionally, the directions and relative magnitudes of the field gradient principal axes are found to be sensitive to the σ and π electron donation by the ligands and the model of the environment. Chlorine-35 NQR spectra of 235 U-depleted and 235 U-enriched Cs 2 UO 2 Cl 4 exhibited no uranium-isotope-dependent shift, but the resonance of the depleted sample displayed a 58% broader line width.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Identification of 2-Hydroxyacyl-CoA Synthases with High Acyloin Condensation Activity for Orthogonal One-Carbon Bioconversion

One-carbon (C1) compounds are emerging as cost-effective and potentially carbon-negative feedstocks for biomanufacturing, which require efficient, versatile metabolic platforms for the synthesis of value-added products. Synthetic formyl-CoA elongation (FORCE) pathways allow diverse product synthesis from C1 compounds via iterative C1 elongation, operating independently from the host metabolism with reduced engineering complexity and improved theoretical yields. However, a major bottleneck was identified as the suboptimal kinetics of the core C1–C1 condensation enzyme, 2-hydroxyacyl-CoA synthase (HACS), catalyzing the acyloin condensation reaction between formaldehyde and formyl-CoA. Furthermore, we used a combinatorial approach of bioprospecting and rational protein engineering to identify multiple HACS variants with significantly improved activities toward C1 substrates. Sequence and structure alignment of the active variants elucidated the key regions for the catalytic function, which were targeted for mutagenesis, leading to improved catalytic efficiency. In parallel, a consecutive round of bioprospecting for homologs with high similarity with active variants revealed a highly active HACS variant exhibiting up to 7-fold improvement in catalytic efficiency (k cat /K M ) and 14-fold improvement in the FORCE pathway flux in vivo compared to the previous reports. Upon further optimization of the downstream pathway, the orthogonal C1-to-product bioconversion system showed a metabolic flux of up to 700 μM glycolate OD –1 h –1 (2.1 mmol gDCW –1 h –1 ) and an industrially relevant glycolate titer, rate, and yield of 5.2 g L –1 (67.8 mM), 0.22 g L –1 h –1 , and 94% carbon yield, respectively.

2-hydroxyacyl-CoA synthase↗
DOE OSTI2022

Avoiding excess computation in asynchronous evolutionary algorithms

Abstract Asynchronous evolutionary algorithms are becoming increasingly popular as a means of making full use of many processors while solving computationally expensive search and optimization problems. These algorithms excel at keeping large clusters fully utilized, but may sometimes inefficiently sample an excess of fast‐evaluating solutions at the expense of higher‐quality, slow‐evaluating ones. We have previously introduced a steady‐state parent selection strategy, SWEET (“Selection whilE EvaluaTing”), that sometimes selects individuals that are still being evaluated and allows them to reproduce early. We perform a takeover‐time analysis that confirms that this strategy gives slow‐evaluating individuals that have higher fitnesses an increased ability to multiply in the population. We also find that SWEET appears effective at improving optimization performance on problems in which solution quality is positively correlated with evaluation time. We evaluate our approach on six simulated real‐valued optimization problems and three real‐world applications: an autonomous vehicle controller problem that involves tuning a spiking neural network and two adversarial EA problems. We further evaluate SWEET versus a basic asynchronous process in a simulated setting. We present evidence that SWEET outperforms basic asynchronous processes in a use‐case in which performance is positively correlated with evaluation time, and performs comparably (and often better) than basic asynchronous processes in several use‐cases where performance is negatively correlated with evaluation time. That said, in the cases where performance and evaluation time are negatively correlated the variance of outcomes for SWEET is notably high.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

NWChem: Past, present, and future

Specialized computational chemistry packages have permanently reshaped the landscape of chemical sciences by providing tools to support and guide the experimental effort and for prediction of chemical and materials properties. In this regard, a special role has been played by electronic structure packages where complex chemical and materials processes can be modeled using first-principle-driven methodologies. Over the last few decades, the rapid development of computing technologies and a tremendous increase in computational power has offered a unique chance to study complex chemical transformations using sophisticated and predictive many-body techniques to describe correlated behavior of electrons in molecular and condensed phase systems at different levels of theory. In enabling these simulations, a critical role has been played by novel parallel algorithms capable of taking advantage of computational resources to address polynomial scaling of electronic structure methods. NWChem was among the first electronic structure codes that focused on delivering scalable performance for electronic structure simulations. Herein, we briefly review the NWChem suite of computational codes including its history, design principles, parallel tools, current capabilities, outreach and outlook.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Positive Neutrino Masses with DESI DR2 via Matter Conversion to Dark Energy

The Dark Energy Spectroscopic Instrument (DESI) is a massively parallel spectroscopic survey on the Mayall telescope at Kitt Peak, which has released measurements of baryon acoustic oscillations determined from over 14 million extragalactic targets. We combine DESI Data Release 2 with CMB datasets to search for evidence of matter conversion to dark energy (DE), focusing on a scenario mediated by stellar collapse to cosmologically coupled black holes (CCBHs). In this physical model, which has the same number of free parameters as Λ⁢CDM, DE production is determined by the cosmic star formation rate density (SFRD), allowing for distinct early- and late-time cosmologies. Using two SFRDs to bracket current observations, we find that the CCBH model: accurately recovers the cosmological expansion history, agrees with early-time baryon abundance measured by BBN, reduces tension with the local distance ladder, and relaxes constraints on the summed neutrino mass ∑𝑚 𝜈 . For these SFRDs, we find a peaked positive ∑𝑚 𝜈 < 0.149 eV (95% confidence) and ∑𝑚 𝜈 = 0.106$^{+0.050}_{−0.069}$ eV, respectively, in good agreement with lower limits from neutrino oscillation experiments. A peak in ∑𝑚 𝜈 > 0 results from late-time baryon consumption in the CCBH scenario and is expected to be a general feature of any model that converts sufficient matter to dark energy during and after reionization.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2026

HydraGNN_Predictive_GFM_2026 - Ensemble of predictive graph foundation models for atomistic materials modeling

This release contains data and parameters of HydraGNN-based graph foundation models trained as a result of the work published in the pre-print "Exascale Multi-Task Graph Foundation Models for Imbalanced, Multi-Fidelity Atomistic Data" by M. Lupo Pasini et al. (https://arxiv.org/abs/2604.15380). We jointly train on 16 open first-principles datasets (544+ million structures covering 85+ elements) using a multi-task architecture with per-dataset heads and a scalable ADIOS2/DDStore data pipeline. On Frontier, we execute six large-scale DeepHyper hyperparameter optimization campaigns in FP64 and promote the top-performing message-passing models to sustained 2,048-node training, yielding a PaiNN-based lead model. The version of HydraGNN used to generate the outputs provided in this release is HydraGNN v5.0 (https://github.com/ORNL/HydraGNN/releases/tag/v5.0) The list of datasets used for the training of the graph foundation model is the following: 1) Alexandria [1] 2) ANI1x [2] 3) MPTrj [3] 4) Open Catalyst 2020 (OC20) [4] 5) Open Catalyst 2022 (OC22) [5] 6) Open Catalyst 2025 (OC25) [6] 7) Open Direct ir Capture 2023 (ODAC23) [7] 8) Open Materials 2024 (OMat24) [8] 9) Open Molecules 2025 (OMol25) [9] 10) OMol25-neutral (subset of OMol25 that contains only molecules with zero total charge) 11) OMol25-non-neutral (subset of OMol25 that contains only molecules with non-zero total charge) 12) Open Polymers 2026 (OPoly2026) [10] 13) Nabla2DFT [11] 14) QCML [12] 15) QM7X [reference 13] 16) transition1x [14] Dataset references: [1] J. Schmidt et al., “A dataset of 175k stable and metastable materials calculated with the PBEsol and SCAN functionals,” Scientific Data, vol. 9, p. 64, 2022. [2] J. S. Smith et al., “The ANI-1ccx and ANI-1x data sets, coupled-cluster and density functional theory properties for molecules,” Scientific Data, vol. 7, p. 134, 2020. [Online]. Available: https: //www.nature.com/articles/s41597-020-0473-z [3] A. Jain et al., “Commentary: The Materials Project: A materials genome approach to accelerating materials innovation,” APL Materials, vol. 1, no. 1, p. 011002, 07 2013. [Online]. Available: https://doi.org/10.1063/1.4812323 [4] L. Chanussot et al., “Open catalyst 2020 (oc20) dataset and community challenges,” ACS Catalysis, vol. 11, no. 10, pp. 6059–6072, 2021. [Online]. Available: https://doi.org/10.1021/acscatal.0c04525 [5] K. Tran et al., “Open catalyst 2022 (oc22) dataset and challenges for oxidation electrocatalysts,” ACS Catalysis, vol. 13, no. 5, pp. 3066–3084, 2023. [Online]. Available: https://doi.org/10.1021/acscatal.2c05426 [6] S. J. Sahoo et al., “The open catalyst 2025 (oc25) dataset and models for solid-liquid interfaces,” arXiv preprint arXiv:2509.17862, 2025. [Online]. Available: https://arxiv.org/abs/2509.17862 [7] A. Sriram et al., “The open DAC 2023 dataset and challenges for sorbent discovery in direct air capture,” ACS Central Science, vol. 10, no. 5, pp. 923–941, 2024. [8] L. Barroso-Luque et al., “Open materials 2024 (omat24) inorganic materials dataset and models,” 2024. [Online]. Available: https://arxiv.org/abs/2410.12771 [9] D. S. Levine et al., “The open molecules 2025 (OMol25) dataset, evaluations, and models,” 2025. [Online]. Available: https://arxiv.org/abs/2505.08762 [10] D. S. Levine et al., The open polymers 2026 (OPoly26) dataset and evaluations,” arXiv preprint arXiv:2512.23117, 2025. [Online]. Available: https://arxiv.org/abs/2512.23117 [11] K. Khrabrov et al., “Nabla2dft: A universal quantum chemistry dataset of drug-like molecules and a benchmark for neural network potentials,” in NeurIPS 2024 Datasets and Benchmarks Track, 2024. [Online]. Available: https://openreview.net/forum?id=ElUrNM9U8c [12] S. Ganscha et al., “The QCML dataset, quantum chemistry reference data from 33.5M DFT and 14.7B semi-empirical calculations,” Scientific Data, vol. 12, p. 406, 2025. [13] J. Hoja et al., “QM7-X, a comprehensive dataset of quantum-mechanical properties spanning the chemical space of small organic molecules,” Scientific Data, vol. 8, p. 43, 2021. [Online]. Available: https://www.nature.com/articles/s41597-021-00812-2 [14] M. Schreiner et al., “Transition1x - a dataset for building generalizable reactive machine learning potentials,” Scientific Data, vol. 9, p. 779, 2022. The folder "datasets_ADIOS2_format" contains the set of pre-processed datasets in Adaptable I/O System (ADIOS) format (https://www.exascaleproject.org/research-project/adios/) that have been used for the development and training of GFMs in this work. The "datasets_ADIOS2_format" directory contains 2 sub-directories, one for the version "v1" of the datasets and one for the version "v2" of the datasets. The version "v1" of the datasets provides values of the total energy as they are extracted from the original data as it was released by the respective institutions. The version "v2" of the datasets provides values of the energy that have been realigned. The realignment was performed by training a linear regression model that predicts the total energy as a function of the chemical composition of the atomistic structure, and then subtract such prediction from the original value of the total energy. Both folders "v1" and "v2" contain 16 sub-directories, each corresponding to an ADIOS2-formatted dataset The folder "DeepHyper-results" contains the configurational files and model's parameters for all the 186 HPO trials that were successfully completed by the scalable hyperparameter optimization (HPO) runs on Frontier. The content of the folder "DeepHyper-results" I structured as follows: 1) task-list.txt: list of mpnn name, jobid, and deephyper task id 2) gfm_${MPNN}_${JOBID}_0.${TASKID}: run directory with checkpoint files 3) gfm_${MPNN}: deephyper summary directory (*.csv) for each specific MPNN type 4) deephyper-experiment-${JOBID}: output and error logs for each job The file "deephyper-sorted.csv" contains the details of each HydraGNN model built and tested by HPO, obtained by merging the (*.csv) filed from each HPO run executed. Out of all the HPO trials, we selected 10 to continue the training of the respective HydraGNN models. Due to limited computational budget available in the LRN070 allocation we could not complete the training till convergence for all these 10 selected models. The folder "models" contains multiple sub-folders, one per each HydraGNN model trained. Each model sub-folder contains the parameters of each HydraGNN model, with multiple checkpoint-restarts. The list of sub-folders are as follows: 1) multidataset_hpo-BEST1-fp64 2) multidataset_hpo-BEST2-fp64 3) multidataset_hpo-BEST3-fp64 4) multidataset_hpo-BEST4-fp64 5) multidataset_hpo-BEST5-fp64 6) multidataset_hpo-BEST6-fp64 7) multidataset_hpo-BEST7-fp64 8) multidataset_hpo-BEST8-fp64 9) multidataset_hpo-BEST9-fp64 10) multidataset_hpo-BEST10-fp64 Within each one of these folders, additional auxiliary log files are provided with descriptions about how the training proceeded. The lead PaiNN-model is contained inside "multidataset_hpo-BEST6-fp64". The file "mlp_branch_weights" contains the parameters of the multi-layer perceptron (MLP) used to reconcile the predictions of the 16 output decoding heads of the HydragNN architectures. The MLP takes in input the chemical composition of the atomistic structure and predicts averaging weights to linearly mix the predictions of each output decoding head toward consolidating them into a single one. The folder "1.1billion-structure-inference" contains 1.1 billion atomistic structures randomly generated. Each structures is associated with energy and forces predicted with the lead-PaiNN model combined with the MLP model for reconciliation of the multi-branch predictions generated by the 16 output decoding heads. The folder "1.1billion-structure-inference" contains 9,300 (*.tar.gz) subdirectories, one per Frontier compute node used to execute the inference at exascale. Once uncompressed, each (*.tar.gz) subdirectory contains an ADIOS2 (*.bp) file container, where each atomistic structure is stored as a PyTorch-Geometric Data object. The file "export_dataset_environment_variables.sh" contains the environment variables that need to be set before running the HydraGNN code to reproduce the results provided in this dataset release. The code that can be used to load the ADIOS2 files, load HydraGNN models, and run inference is available at: https://github.com/ORNL/HydraGNN/releases/tag/v5.0

36 MATERIALS SCIENCE↗