Search NASA⌕ Search

SEARCH · Search NASA

Results for “Scalable Model”

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.

At least 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Distributed-Memory Sparse Deep Neural Network Inference Using Global Arrays

Partitioned Global Address Space (PGAS) models exhibit tremendous promise in developing efficient and productive distributed-memory parallel applications. They have been used extensively in scientific computations due to conveniently offering a ``shared-memory''-like model and convenient interfaces that separate communication with synchronization. Traditionally, PGAS communication models have been applied to dense/contiguously distributed data, but most modern applications depict varied levels of sparsity. Existing PGAS models require certain adaptations to support distributed sparse computations, since associated computations often require matrix arithmetic, in addition to data movement. The Global Arrays toolkit from Pacific Northwest National Laboratory (PNNL) is one of the earliest PGAS models to combine one-sided data communication and distributed matrix operations and is still used in the popular NWChem quantum chemistry suite. Recently, we have expanded the Global Arrays toolkit to support common sparse operations, like sparse matrix-dense matrix multiplies (SpMM), sparse matrix-sparse matrix multiplication (SpGEMM) and Sampled Dense-Dense Matrix Multiplication (SDDMM). As it turns out, these operations are the bedrock of sparse Deep Learning (DL); sparse deep neural networks and Graph Neural Networks (GNNs) have gained increasing attention recently in achieving speedups on training and inference with reduced memory footprints. Unlike scientific applications in High Performance Computing (HPC), modern (distributed-memory capable) DL toolkits often rely on non-standardized and closed-source vendor software optimizations, creating challenges in software-hardware co-design at scale. Our goal is to support a variety of distributed-memory sparse matrix operations and helper functions in the newly created Sparse Global Arrays (SGA), such that it is possible to build portable and productive Machine Learning scenarios for algorithm/software and hardware codesign purposes. Contemporary data-parallel schemes for training/inference are undergoing a major overhaul since model replication limits scalability and causes resource inefficiencies. As such, we have adopted tensor parallelism in decomposing the model and inputs, to mitigate memory issues. Current implementation is built on top of MPI and uses CPUs to maximize the portability across the platforms.

Distributed computing, machine learning↗
DOE OSTI2024

Stochastic Trust-Region Algorithm in Random Subspaces with Convergence and Expected Complexity Analyses

Here, this work proposes a framework for large-scale stochastic derivative-free optimization (DFO) by introducing STARS, a trust-region method based on iterative minimization in random subspaces. This framework is both an algorithmic and theoretical extension of a random subspace derivative-free optimization (RSDFO) framework, and an algorithm for stochastic optimization with random models (STORM). Moreover, like RSDFO, STARS achieves scalability by minimizing interpolation models that approximate the objective in low-dimensional affine subspaces, thus significantly reducing per-iteration costs in terms of function evaluations and yielding strong performance on largescale stochastic DFO problems. The user-determined dimension of these subspaces, when the latter are defined, for example, by the columns of so-called Johnson-Lindenstrauss transforms, turns out to be independent of the dimension of the problem. For convergence purposes, inspired by the analyses of RSDFO and STORM, both a particular quality of the subspace and the accuracies of random function estimates and models are required to hold with sufficiently high, but fixed, probabilities. Using martingale theory under the latter assumptions, an almost sure global convergence of STARS to a first-order stationary point is shown, and the expected number of iterations required to reach a desired first-order accuracy is proved to be similar to that of STORM and other stochastic DFO algorithms, up to constants.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Electrification Analysis: Container Ports' Cargo Handling Equipment

This one-page highlight details the key takeaways from a project that utilized NREL's Fleet Research, Energy Data, and Insights (FleetREDI) data analysis pipeline, the Electrification Analysis of Container Ports' Cargo Handling Equipment project. This project created a scalable solution to model energy demand per shipping container moved (kWh/TEU) for an all-electric cargo handling equipment fleet located at a maritime port. The model allows stakeholders to understand energy demand at each electric vehicle (EV) equipment level and is easily scalable to container demand and EV adoption rate projections.

ADVANCED PROPULSION SYSTEMS↗
DOE OSTI2025

Materials Learning Algorithms (MALA): Scalable machine learning for electronic structure calculations in large-scale atomistic simulations

We present the Materials Learning Algorithms (MALA) package, a scalable machine learning framework designed to accelerate density functional theory (DFT) calculations suitable for large-scale atomistic simulations. Using local descriptors of the atomic environment, MALA models efficiently predict key electronic observables, including local density of states, electronic density, density of states, and total energy. The package integrates data sampling, model training and scalable inference into a unified library, while ensuring compatibility with standard DFT and molecular dynamics codes. We demonstrate MALA's capabilities with examples including boron clusters, aluminum across its solid-liquid phase boundary, and predicting the electronic structure of a stacking fault in a large beryllium slab. Scaling analyses reveal MALA's computational efficiency and identify bottlenecks for future optimization. With its ability to model electronic structures at scales far beyond standard DFT, MALA is well suited for modeling complex material systems, making it a versatile tool for advanced materials research.

Density functional theory↗
DOE OSTI2024

Thermochemical Nonequilibrium Modeling in a Continuous-Galerkin, Finite-Element Framework

The presented work discusses the implementation, verification, and validation of Park's two-temperature model in a scalable, computational fluid dynamics (CFD) code developed at the US Department of Energy's Oak Ridge National Laboratory (ORNL). The implementation of Park's two-temperature model was verified through 0D test cases involving an adiabatic reactor and a nitrogen thermal bath. The implementation was then validated through comparisons with other validated CFD codes and experimental data on a hypersonic cylinder and double cones. These are standard benchmark test cases for thermochemical non-equilibrium (TCNE) modeling, and all data are shared publicly. The verification and validation results showed that ORNL's in-house CFD code could model complex, high-speed flow problems with and without TCNE modeling. This work is essential for future research involving 3D shock wave/boundary layer interactions (SBLIs).

Nutter, Nicole↗
DOE OSTI2026

Designing resilient IoT and Edge Computing with federated tinyML

The rapid growth of the Internet of Things (IoT) and Edge Computing (EC) has brought significant conveniences to modern society but has also greatly expanded the cyber attack surfaces, particularly as these technologies are being increasingly integrated into critical systems such as power grids, healthcare, and smart homes. Here, to improve IoT/EC’s cybersecurity posture, we leveraged Artificial Intelligence (AI) and Machine Learning (ML) by employing tinyML to monitor voluminous IoT data for cyber threats while addressing devices’ resource constraints, and utilizing Federated Learning (FL) to share local detection knowledge across the system while preserving privacy. Building on our three-layer architecture combining tinyML and FL to enhance autonomous cyber attack detection, this paper demonstrated that the architecture improves detection accuracy, reduces resource consumption, and enables lightweight, secure IoT device monitoring. These results were validated using the public N-BaIoT dataset as well as real IoT network traffic data collected under multiple attack scenarios from our testbeds. Additionally, we introduced an enhanced FL methodology with a novel preprocessing stage, including federated feature selection and global preprocessor construction, to address IoT/EC data heterogeneity. We developed a physical IoT testbed for attack simulations and data collection, implemented a tinyML-powered detector for realistic model validation, and also built a virtual testbed for scalable evaluations of FL models across diverse network environments.

Cognitive cyber↗
DOE OSTI2025

Surrogate Constructed Scalable Circuits ADAPT-VQE in the Schwinger model

Inspired by recent advancements of simulating periodic systems on quantum computers, we develop a new approach, (SC)$^2$-ADAPT-VQE, to further advance the simulation of these systems. Our approach extends the scalable circuits ADAPT-VQE framework, which builds an ansatz from a pool of coordinate-invariant operators defined for arbitrarily large, though not arbitrarily small, volumes. Our method uses a classically tractable ``Surrogate Constructed'' method to remove irrelevant operators from the pool, reducing the minimum size for which the scalable circuits are defined. Bringing together the scalable circuits and the surrogate constructed approaches forms the core of the (SC)$^2$ methodology. Our approach allows for a wider set of classical computations, on small volumes, which can be used for a more robust extrapolation protocol. While developed in the context of lattice models, the surrogate construction portion is applicable to a wide variety of problems where information about the relative importance of operators in the pool is available. As an example, we use it to compute properties of the Schwinger model - quantum electrodynamics for a single, massive fermion in $1+1$ dimensions - and show that our method can be used to accurately extrapolate to the continuum limit.

Gustafson, Erik [RIACS, Mtn. View] (ORCID:00000001↗
DOE OSTI2024

A Framework for Identifying Building Energy Models of Localized Utility Service Areas Using Smart Meter Data

Bottom-up load modeling of buildings offers a versatile approach to simulating baseline demand and scenarios of future technology evolution and adoption at the individual building level. This capability is essential to understanding how future load shapes may change with the adoption of electric equipment and vehicles, particularly as it relates to grid planning and infrastructure investments. Traditionally, grid planning techniques have used historical load data to predict future load and infrastructure needs. However, with the anticipated rise in adoption of electrification technologies such as heat pumps and electric vehicles, historical data become less reliable predictors of the future. By employing ResStock, a high-fidelity building stock modeling tool, we can fine-tune electrification scenarios and aggregate models to represent varying geographic resolutions of the grid system, while considering the underlying features of homes. This may enable a more accurate and responsive approach to anticipate and plan for the evolving landscape of energy demands. We present a new framework that leverages building stock energy modeling to identify building models that align with the load shapes and housing attributes of buildings with AMI data. This approach applies two model layers: (1) a classification step that identifies the presence of air conditioning, electric heating, and electric water heating, and (2) an optimization routine that identifies building energy models aligning with load profile data from advanced metering infrastructure meters. This report demonstrates one approach to deploying this framework, and presents results for three test cases that use both modeled and AMI data to assess performance. For a test case using AMI data in Fort Collins, Colorado, we observed a median monthly electricity load CV-RMSE of 16.6%, and a top ten daily heating and cooling median absolute percent error of 7.7% and 8.3%, respectively. For each AMI meter, we identify a set of potential energy models so that downstream use-cases can account for uncertainty driven by variability of baseline technologies and occupant behavior, which impact the response to electrification and energy efficiency scenarios. Our results indicate that ResStock has potential as a scalable solution for modeling residential energy demand at local grid resolutions. Its performance depends on location-specific factors, underlying building characteristics, and the level of aggregation, offering a path towards more precise and adaptive distribution grid planning for the evolving energy landscape.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2025

Scalable training of trustworthy and energy-efficient predictive graph foundation models for atomistic materials modeling: a case study with HydraGNN

We present our work on developing and training scalable, trustworthy, and energy-efficient predictive graph foundation models (GFMs) using HydraGNN, a multi-headed graph convolutional neural network architecture. HydraGNN expands the boundaries of graph neural network (GNN) computations in both training scale and data diversity. It abstracts over message passing algorithms, allowing both reproduction of and comparison across algorithmic innovations that define nearest-neighbor convolution in GNNs. This work discusses a series of optimizations that have allowed scaling up the GFMs training to tens of thousands of GPUs on datasets consisting of hundreds of millions of graphs. Our GFMs use multitask learning (MTL) to simultaneously learn graph-level and node-level properties of atomistic structures, such as energy and atomic forces. Using over 154 million atomistic structures for training, we illustrate the performance of our approach along with the lessons learned on two state-of-the-art US Department of Energy (US-DOE) supercomputers, namely the Perlmutter petascale system at the National Energy Research Scientific Computing Center and the Frontier exascale system at Oak Ridge Leadership Computing Facility. The HydraGNN architecture enables the GFM to achieve near-linear strong scaling performance using more than 2000 GPUs on Perlmutter and 16,000 GPUs on Frontier.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

An information-matching approach to optimal experimental design and active learning

The efficacy of mathematical models heavily depends on the quality of the training data, yet collecting sufficient data is often expensive and challenging. Many modeling applications require inferring parameters only as a means to predict other quantities of interest (QoI). Because models often contain many unidentifiable (sloppy) parameters, QoIs often depend on a relatively small number of parameter combinations. Therefore, we introduce an information-matching criterion based on the Fisher information matrix to select the most informative training data from a candidate pool. This method ensures that the selected data contain sufficient information to learn only those parameters that are needed to constrain downstream QoIs. It is formulated as a convex optimization problem, making it scalable to large models and datasets. Here, we demonstrate the effectiveness of this approach across various modeling problems in diverse scientific fields, including power systems and underwater acoustics. Finally, we use information-matching as a query function within an active learning (AL) loop for materials science applications. In all these applications, we find that a relatively small set of optimal training data can provide the necessary information for achieving precise predictions. These results are encouraging for diverse future applications, particularly AL in large machine-learning models.

Materials science↗
DOE OSTI2025

ThinCurr: An open-source 3D thin-wall eddy current modeling code for the analysis of large-scale systems of conducting structures

In this paper we present a new thin-wall eddy current modeling code, ThinCurr, for studying inductively-coupled currents in 3D conducting structures -- with primary application focused on the interaction between currents flowing in coils, plasma, and conducting structures of magnetically-confined plasma devices. The code utilizes a boundary finite element method on an unstructured, triangular grid to accurately capture device structures. The new code, part of the broader Open FUSION Toolkit, is open-source and designed for ease of use without sacrificing capability and speed through a combination of Python, Fortran, and C/C++ components. Scalability to large models is enabled through use of hierarchical off-diagonal low-rank compression of the inductance matrix, which is otherwise dense. Ease of handling large models of complicated geometry is further supported by automatic determination of supplemental elements through a greedy homology approach. Here, a detailed description of the numerical methods of the code and verification of the implementation of those methods using cross-code comparisons against the VALEN code and Ansys commercial analysis software is shown.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Theoretical Insights into Reaction-Induced Transformation and Tuning of Catalytic Behavior in Heterogenous Catalysis

Reaction-induced transformations in heterogenous catalysis represent diverse phenomena that challenge traditional views of static catalyst surfaces. From surface adsorbate dynamics, atomic rearrangements, to composition and phase transitions, these processes reveal the profound differences between idealized model systems under ultrahigh vacuum and the complex, evolving interfaces that govern real catalytic behaviors under reaction conditions. Here, this perspective reviews recent theoretical efforts to provide atomic-level mechanistic insights into significant reaction-induced transformations and their impact on catalytic activity and selectivity. It underscores the need for an integrated framework that combines predictive simulations with operando characterization to uncover active sites and mechanisms under realistic operating conditions. Achieving this requires accelerating existing simulations to fully capture diverse reaction-induced surface dynamics, enabling scalable and accurate modeling of catalysts as condition-dependent, dynamically evolving systems. Such approaches are critical to bridge the gap between theory and practice, offering a pathway to more impactful and predictive catalyst design.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

A machine learning method of modern urban building energy modeling: A case study of Chicago

Urban-scale building energy modeling is vital for urban planning. However, it can be challenging to assimilate reliable non-geometry building data for urban-scale modeling without extensive investment. Here, this study introduces a novel approach to developing modern urban-scale building energy stock data using geographic information systems and machine learning algorithms without necessarily requiring pre-supplied non-geometric metadata. The proposed framework integrates building footprint and height data to estimate gross floor areas, and matches each building to a pool of candidate records from ComStock or ResStock—filtered to the same county and ranked by geometric similarity—demonstrate a proof-of-concept case study in Chicago for predicting energy use intensity (EUI) using scalable datasets. The model achieved a mean bias error (MBE) of 0.08 kWh/m² and root mean square error (RMSE) of 14.84 kWh/m² under full metadata input for EUI prediction. With only location inputs, the model captured 69.2 % of EUI within predicted ranges. These results demonstrate the model’s potential to support early-stage urban planning, identify candidates for energy-efficient retrofits. By removing the dependency on detailed pre-surveys or extensive building metadata, the approach overcomes a key barrier in traditional urban-scale building energy modeling, illustrating a pathway toward broader and more cost-effective application, though further multi-city validation and improved treatment of pre-1925 buildings are needed.

Energy Use Intensity↗
DOE OSTI2025

Preparing MPICH for exascale

The advent of exascale supercomputers heralds a new era of scientific discovery, yet it introduces significant architectural challenges that must be overcome for MPI applications to fully exploit its potential. Among these challenges is the adoption of heterogeneous architectures, particularly the integration of GPUs to accelerate computation. Additionally, the complexity of multithreaded programming models has also become a critical factor in achieving performance at scale. The efficient utilization of hardware acceleration for communication, provided by modern NICs, is also essential for achieving low latency and high throughput communication in such complex systems. In response to these challenges, the MPICH library, a high-performance and widely used Message Passing Interface (MPI) implementation, has undergone significant enhancements. Here, this paper presents four major contributions that prepare MPICH for the exascale transition. First, we describe a lightweight communication stack that leverages the advanced features of modern NICs to maximize hardware acceleration. Second, our work showcases a highly scalable multithreaded communication model that addresses the complexities of concurrent environments. Third, we introduce GPU-aware communication capabilities that optimize data movement in GPU-integrated systems. Finally, we present a new datatype engine aimed at accelerating the use of MPI derived datatypes on GPUs. These improvements in the MPICH library not only address the immediate needs of exascale computing architectures but also set a foundation for exploiting future innovations in high-performance computing. By embracing these new designs and approaches, MPICH-derived libraries from HPE Cray and Intel were able to achieve real exascale performance on OLCF Frontier and ALCF Aurora respectively.

Guo, Yanfei [Argonne National Laboratory (ANL), Ar↗
DOE OSTI2025

Surrogate-constructed scalable-circuits adaptive variational quantum eigensolver in the Schwinger model

Inspired by recent advancements in simulating periodic systems on quantum computers, we develop an approach to further advance the simulation of these systems, named (SC) 2 -ADAPT-VQE. Our approach extends the scalable-circuits ADAPT-VQE framework, which builds an ansatz from a pool of coordinate-invariant operators defined for arbitrarily large, though not arbitrarily small, volumes. Our method uses a classically tractable “surrogate constructed” method to remove irrelevant operators from the pool, reducing the minimum size for which the scalable circuits are defined. Bringing together the scalable circuits and the surrogate constructed approaches forms the core of the (SC) 2 methodology. Our approach allows for a wider set of classical computations on small volumes, which can be used for a more robust extrapolation protocol. While developed in the context of lattice models, the surrogate construction portion is applicable to a wide variety of problems where information about the relative importance of operators in the pool is available. As an example, we use it to compute the properties of the Schwinger model—quantum electrodynamics for a single, massive fermion in 1 +1 dimensions—and show that our method can be used to accurately extrapolate to the continuum limit.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Scalable workflow for evaluating and optimizing large language models

This work describes the improved workflow for evaluating open-source large language models (LLMs) for trustworthiness. The workflow facilitates the acquisition of LLMs, the generation of LLM responses, and the evaluation of the responses for their trustworthiness. As a use case, the workflow is employed to evaluate dense, quantized, and pruned Meta Llama3.1 LLMs for their truthfulness. The outcome of the project could set the stage for understanding and developing trustworthy models in the future projects.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Integrating Ultra-Coarse-Grained Protein Models into Accessible Workflows for Multiscale Molecular Dynamics

To capture protein conformational transitions using molecular dynamics (MD), several simulation resolutions covering different spatial and temporal scales are typically needed. All-atom (AA) simulations provide fine resolution, but are computationally infeasible for large systems over longer durations. Coarse-grained (CG) and ultra-coarse-grained (UCG) models have a lower resolution and computational cost while still being able to conserve essential protein features. Prior work on a Multiscale Machinelearned Modeling Infrastructure (MuMMI) combined both AA and CG simulations to study RAS-RAF protein interactions, leveraging CG models for longer time scales and using AA to investigate unusual conformations in greater detail. However, MuMMI is still resource-intensive, and this study aims to maximize exploration of the protein conformational space while reducing computational cost. In this paper, we build on prior work that integrates UCG models based on heterogeneous elastic network modeling (hENM) into the MuMMI workflow. We demonstrate that UCG models enable accurate sampling of protein conformations, focusing on simulating RAS-RAF protein interactions. Using higher-resolution CG Martini simulation data, we can automatically refine intramolecular interactions in UCG models. We present a scalable Python package that uses fluctuations observed in higher-resolution CG Martini simulations to estimate bond coefficients of the UCG model. We built novel machine learning-based backmapping methods to recover more detailed CG Martini structures from UCG structures, using diffusion models to learn the mapping between scales. Finally, we present UCG-mini-MuMMI, an accessible and less compute-intensive version of MuMMI as a resource for the scientific community. Incorporating UCG models into MD studies is applicable to a broad range of systems and proteins, and our study offers insights into the advantages and limitations of these methods.

Chemical structure↗
DOE OSTI2026

A Co-Simulation Framework for Steady-State Analyses of Multiple Droop-based MTdc Grids in Continental-Scale Systems

The growing scale and complexity of planning continental hybrid ac and multi-terminal dc (MTdc) systems require scalable steady-state modeling and analysis approaches not currently available in commercial tools. This paper presents a comprehensive multi-fidelity model-conversion framework that enables the efficient transition of MTdc grid models from production cost modeling (PCM) and approximated ac power flow to detailed ac–MTdc power flow for large-scale planning studies. The core of this framework is a scalable co-simulation approach that, for the first time, enables power flow analysis in continental-scale ac–MTdc systems. It seamlessly couples commercial ac solvers with a detailed MTdc grid model that incorporates droop-based control and current-limiting strategies of multiple meshed MTdc grids. Leveraging this capability, an evaluation framework to systematically assess and compare different MTdc power redispatch strategies under ac and dc contingencies is introduced. The proposed framework and algorithm are evaluated using a combined Western and Eastern Interconnection system with 11 MTdc grids of various sizes, showing a coherent transition from PCM to detailed ac–MTdc power flow and improved system performance in voltage regulation and line overload mitigation following typical contingencies.

Nguyen, Quan H.↗