SEARCH · Search NASA

Results for “Architecture patterns”

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 73 records · Page 4

All sources NASA NTRS DOE OSTI

DOE OSTI2024

Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting

Advancing the capabilities of earthquake nowcasting, the real-time forecasting of seismic activities, remains crucial for reducing casualties. This multifaceted challenge has recently gained attention within the deep learning domain, facilitated by the availability of extensive earthquake datasets. Despite significant advancements, the existing literature on earthquake nowcasting lacks comprehensive evaluations of pre-trained foundation models and modern deep learning architectures; each focuses on a different aspect of data, such as spatial relationships, temporal patterns, and multi-scale dependencies. This paper addresses the mentioned gap by analyzing different architectures and introducing two innovative approaches called Multi Foundation Quake and GNNCoder. We formulate earthquake nowcasting as a time series forecasting problem for the next 14 days within 0.1-degree spatial bins in Southern California. Earthquake time series are generated using the logarithm energy released by quakes, spanning 1986 to 2024. Our comprehensive evaluations demonstrate that our introduced models outperform other custom architectures by effectively capturing temporal-spatial relationships inherent in seismic data. The performance of existing foundation models varies significantly based on the pre-training datasets, emphasizing the need for careful dataset selection. However, we introduce a novel method, Multi Foundation Quake, that achieves the best overall performance by combining a bespoke pattern with Foundation model results handled as auxiliary streams.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Understanding and Leveraging the I/O Patterns of Emerging Machine Learning Analytics

The scientific community is currently experiencing unprecedented amounts of data generated by cutting-edge science facilities. Soon facilities will be producing up to 1 PB/s which will force scientist to use more autonomous techniques to learn from the data. The adoption of machine learning methods, like deep learning techniques, in large-scale workflows comes with a shift in the workflow’s computational and I/O patterns. These changes often include iterative processes and model architecture searches, in which datasets are analyzed multiple times in different formats with different model configurations in order to find accurate, reliable and efficient learning models. This shift in behavior brings changes in I/O patterns at the application level as well at the system level. These changes also bring new challenges for the HPC I/O teams, since these patterns contain more complex I/O workloads. In this paper we discuss the I/O patterns experienced by emerging analytical codes that rely on machine learning algorithms and highlight the challenges in designing efficient I/O transfers for such workflows. We comment on how to leverage the data access patterns in order to fetch in a more efficient way the required input data in the format and order given by the needs of the application and how to optimize the data path between collaborative processes. We will motivate our work and show performance gains with a study case of medical applications.

Gainaru, Ana↗

DOE OSTI2026

HARMONY: Large-Scale Architecture Search for Efficient Hybrid Language Models

As large language models scale to trillions of parameters, their computational and memory requirements present critical challenges for efficient training and deployment. While Mixture of Experts (MoE) architectures enable efficient scaling through sparse parameter activation, and state-space models like Mamba offer linear-time complexity, principled methods for combining these paradigms remain undeveloped. We introduce HARMONY (Hybrid Architecture Research for Mamba, Optimized with Neural efficiencY), a multi-objective evolutionary neural architecture search framework for discovering efficient hybrid language models that integrate Transformer attention mechanisms, Mixture-of-Experts routing, and Mamba state-space components. Through large-scale distributed search using 16,384 MI250X GPUs on the Frontier supercomputer, HARMONY explores a comprehensive design space encompassing six attention variants (MHA, MQA, GQA, MLA, SWA, and Mamba-2), variable MoE configurations with both routed and shared experts, and extensive Mamba hyperparameters. Our framework discovers heterogeneous architectures that balance training performance with computational efficiency through multi-objective optimization incorporating latency penalties and fitness-based selection. Analysis of discovered architectures reveals that optimal hybrid designs favor heterogeneous component mixing rather than homogeneous patterns, with Mamba-2 and Multi-Head Latent Attention (MLA) emerging as preferred mechanisms. Discovered architectures demonstrate superior training efficiency: our best configuration achieves a final perplexity of 1.0874 with 2.38B parameters while processing 4,320 tokens/second, outperforming significantly larger manually designed models. Full-scale evaluation shows HARMONY's top architectures achieve better loss trajectories than equivalently-sized models using state-of-the-art configurations including Mixtral, Jamba, and Samba. Additionally, we demonstrate 91% weak scaling efficiency when training discovered 36B-parameter models across 1,024 GPUs. HARMONY is released as an open framework with comprehensive tools for building and training hybrid models using expert-data-pipeline parallelism, democratizing access to automated architecture design for next-generation language models.

Herron, Emily [ORNL] (ORCID:0000000273008172)↗

DOE OSTI2020

Effects of fabric architectures on mechanical and damage behaviors in carbon/epoxy woven composites under multiaxial stress states

The mechanical properties and damage evolutions of carbon/epoxy woven fabric composites with three different fabric architectures, including one plain weave and two twill weave patterns, are experimentally investigated under multiaxial stress states. In particular, the effects of weave patterns are investigated by monotonic and cyclic off-axis tension tests. Both elastic modulus and strength degrade remarkably with increasing off-axis loading angle, while Poisson's ratio is much higher than that measured from on-axis tests and increases with loading strain gradually. Different fabric architectures show limited effects on the modulus and strength under multiaxial stress states, and they are well predicted by transformation equation and Tsai-Wu failure criteria, respectively. However, significantly different failure behaviors are observed in three fabric composites, and microstructure observation shows that fabric architecture affects the stress concentration and the damage development. Smaller crimp ratio and compacted structure postpone the damage development but result in more abrupt failure under multiaxial stress states.

36 MATERIALS SCIENCE↗

DOE OSTI2022

Polymeric multimaterials by photochemical patterning of crystallinity

An organized combination of stiff and elastic domains within a single material can synergistically tailor bulk mechanical properties. However, synthetic methods to achieve such sophisticated architectures remain elusive. We report a rapid, facile, and environmentally benign method to pattern strong and stiff semicrystalline phases within soft and elastic matrices using stereo-controlled ring-opening metathesis polymerization of an industrial monomer, cis -cyclooctene. Dual polymerization catalysis dictates polyolefin backbone chemistry, which enables patterning of compositionally uniform materials with seamless stiff and elastic interfaces. Visible light–induced activation of a metathesis catalyst results in the formation of semicrystalline trans polyoctenamer rubber, outcompeting the formation of cis polyoctenamer rubber, which occurs at room temperature. This bottom-up approach provides a method for manufacturing polymeric materials with promising applications in soft optoelectronics and robotics.

Science & Technology - Other Topics↗

DOE OSTI2020

Experiences in porting mini-applications to OpenACC and OpenMP on heterogeneous systems

This article studies mini-applications—Minisweep, GenASiS , GPP, and FF—that use computational methods commonly encountered in HPC. We have ported these applications to develop OpenACC and OpenMP versions, and evaluated their performance on Titan (Cray XK7 with K20x GPUs), Cori (Cray XC40 with Intel KNL), Summit (IBM AC922 with Volta GPUs), and Cori-GPU (Cray CS-Storm 500NX with Intel Skylake and Volta GPUs). Our goals are for these new ports to be useful to both application and compiler developers, to document and describe the lessons learned and the methodology to create optimized OpenMP and OpenACC versions, and to provide a description of possible migration paths between the two specifications. Cases where specific directives or code patterns result in improved performance for a given architecture are highlighted. Here, we also include discussions of the functionality and maturity of the latest compilers available on the above platforms with respect to OpenACC or OpenMP implementations.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2020

The parallelism motifs of genomic data analysis

Genomic datasets are growing dramatically as the cost of sequencing continues to decline and small sequencing devices become available. Enormous community databases store and share these data with the research community, but some of these genomic data analysis problems require large-scale computational platforms to meet both the memory and computational requirements. These applications differ from scientific simulations that dominate the workload on high-end parallel systems today and place different requirements on programming support, software libraries and parallel architectural design. For example, they involve irregular communication patterns such as asynchronous updates to shared data structures. We consider several problems in high-performance genomics analysis, including alignment, profiling, clustering and assembly for both single genomes and metagenomes. We identify some of the common computational patterns or ‘motifs’ that help inform parallelization strategies and compare our motifs to some of the established lists, arguing that at least two key patterns, sorting and hashing, are missing. This article is part of a discussion meeting issue ‘Numerical algorithms for high-performance computational science’.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2026

Interpretable Models for Workflow Differentiation in High-Performance Scientific Networks

Scientific workflows in high-performance networks spawn hundreds of interdependent flows that must be managed collectively—yet existing network classifiers treat each flow in isolation, leading to fragmented QoS decisions and missed interflow patterns. We present a novel traffic classification solution that operates at the workflow level, distinguishing entire filetransfer operations from streaming analytics by capturing how concurrent flows interact and burst together. We introduce a workflow identification window (WIW) that ingests raw packet headers from parallel flows into unified tensors, preserving the spatial-temporal patterns that differentiate scientific workflows. This approach achieves 98.7% accuracy using CNN, LSTM, and hybrid architectures, while maintaining 84% accuracy on production traffic collected a week later—demonstrating robustness to temporal drift. By integrating SHAP and GradCAM explainability, we reveal that early-packet timing patterns and cross-flow correlations drive classification decisions, providing operators with interpretable insights. Our system enables coherent workflow-level QoS enforcement and dynamic bandwidth allocation in scientific networks, eliminating manual per-flow configuration while maintaining classification latency at millisecond level.

Giannakou, Anna [LBL, Berkeley]↗

DOE OSTI2020

Future Vision for Autonomous Ocean Observations

Autonomous platforms already make observations over a wide range of temporal and spatial scales, measuring salinity, temperature, nitrate, pressure, oxygen, biomass; and many other parameters. However, the observations are not comprehensive. Future autonomous systems need to be more affordable, more modular, more capable and easier to operate. Creative new types of platforms and new compact, low power, calibrated and stable sensors are under development to expand autonomous observations. Communications and recharging need bandwidth and power which can be supplied by standardized docking stations. In situ power generation will also extend endurance for many types of autonomous platforms, particularly autonomous surface vehicles. Standardized communications will improve ease of use, interoperability, and enable coordinated behaviors. Improved autonomy and communications will enable adaptive networks of autonomous platforms. Improvements in autonomy will have three aspects: hardware, control, and operations. As sensors and platforms have more onboard processing capability and energy capacity, more measurements become possible. Control systems and software will have the capability to address more complex states and sophisticated reactions to sensor inputs, which allows the platform to handle a wider variety of circumstances without direct operator control. Operational autonomy is increased by reducing operating costs. To maximize the potential of autonomous observations new standards and best practices are needed. In some applications, focus on common platforms and volume purchases could lead to significant cost reductions. Cost reductions could enable order-of-magnitude increases in platform operations and increase sampling resolution for a given level of investment. Energy harvesting technologies should be integral to the system design, for sensors, platforms, vehicles, and docking stations. Connections are needed between the marine energy and ocean observing communities to coordinate among funding sources, researchers, and end users. Regional teams should work with global organizations such as IOC/GOOS in governance development. International networks such as EGO for the emerging glider operations should also provide a forum for addressing governance. Networks of multiple vehicles can improve operational efficiencies and transform operational patterns. There is a need to develop operational architectures at regional and global scales to provide a backbone for active networking of autonomous platforms.

47 OTHER INSTRUMENTATION↗

DOE OSTI2021

Hardware-Based Emulator with Deep Learning Model for Building Energy Control and Prediction Based on Occupancy Sensors’ Data

Heating, ventilation, and air conditioning (HVAC) is the largest source of residential energy consumption. Occupancy sensors’ data can be used for HVAC control since it indicates the number of people in the building. HVAC and sensors form a typical cyber-physical system (CPS). In this paper, we aim to build a hardware-based emulation platform to study the occupancy data’s features, which can be further extracted by using machine learning models. In particular, we propose two hardware-based emulators to investigate the use of wired/wireless communication interfaces for occupancy sensor-based building CPS control, and the use of deep learning to predict the building energy consumption with the sensor data. We hypothesize is that the building energy consumption may be predicted by using the occupancy data collected by the sensors, and question what type of prediction model should be used to accurately predict the energy load. Another hypothesis is that an in-lab hardware/software platform could be built to emulate the occupancy sensing process. The machine learning algorithms can then be used to analyze the energy load based on the sensing data. To test the emulator, the occupancy data from the sensors is used to predict energy consumption. The synchronization scheme between sensors and the HVAC server will be discussed. We have built two hardware/software emulation platforms to investigate the sensor/HVAC integration strategies, and used an enhanced deep learning model—which has sequence-to-sequence long short-term memory (Seq2Seq LSTM)—with an attention model to predict the building energy consumption with the preservation of the intrinsic patterns. Because the long-range temporal dependencies are captured, the Seq2Seq models may provide a higher accuracy by using LSTM architectures with encoder and decoder. Meanwhile, LSTMs can capture the temporal and spatial patterns of time series data. The attention model can highlight the most relevant input information in the energy prediction by allocating the attention weights. The communication overhead between the sensors and the HVAC control server can also be alleviated via the attention mechanism, which can automatically ignore the irrelevant information and amplify the relevant information during CNN training. Our experiments and performance analysis show that, compared with the traditional LSTM neural network, the performance of the proposed method has a 30% higher prediction accuracy.

Ye, Zhijing↗

DOE OSTI2020

Three-dimensional deterministic graphene architectures formed using three-dimensional templates

In one embodiment, a composition of matter includes: a plurality of ligaments each independently comprising one or more layers of graphene; where the plurality of ligaments are arranged according to a deterministic three-dimensional (3D) pattern. In another embodiment, a method of forming a deterministic three-dimensional (3D) architecture of graphene includes: forming or providing a substrate structurally characterized by a predefined 3D pattern; forming one or more layers of metal on surfaces of the substrate; and forming one or more layers of graphene on surfaces of the metal.

36 MATERIALS SCIENCE↗

DOE OSTI2024

Architecture and performance of Perlmutter's 35 PB ClusterStor E1000 all-flash file system

NERSC's newest system, Perlmutter, features a 35 PB all-flash Lustre file system built on HPE Cray ClusterStor E1000. Here, we present its architecture, early performance figures, and performance considerations unique to this architecture. We demonstrate the performance of E1000 OSSes through low-level Lustre tests that achieve over 90% of the theoretical bandwidth of the SSDs at the OST and LNet levels. We also show end-to-end performance for both traditional dimensions of I/O performance (peak bulk-synchronous bandwidth) and nonoptimal workloads endemic to production computing (small, incoherent I/Os at random offsets) and compare them to NERSC's previous system, Cori, to illustrate that Perlmutter achieves the performance of a burst buffer and the resilience of a scratch file system. Finally, we discuss performance considerations unique to all-flash Lustre and present ways in which users and HPC facilities can adjust their I/O patterns and operations to make optimal use of such architectures.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2021

Architecture and Performance of Perlmutter’s 35 PB ClusterStor E1000 All-Flash File System

NERSC's newest system, Perlmutter, features a 35 PB all-flash Lustre file system built on HPE Cray ClusterStor E1000. We present its architecture, early performance figures, and performance considerations unique to this architecture. We demonstrate the performance of E1000 OSSes through low-level Lustre tests that achieve over 90% of the theoretical bandwidth of the SSDs at the OST and LNet levels. We also show end-to-end performance for both traditional dimensions of I/O performance (peak bulk-synchronous bandwidth) and non-optimal workloads endemic to production computing (small, incoherent I/Os at random offsets) and compare them to NERSC's previous system, Cori, to illustrate that Perlmutter achieves the performance of a burst buffer and the resilience of a scratch file system. Finally, we discuss performance considerations unique to all-flash Lustre and present ways in which users and HPC facilities can adjust their I/O patterns and operations to make optimal use of such architectures.

Lockwood, Glenn K↗

DOE OSTI2024

Accurate and Data‐Efficient Micro X‐ray Diffraction Phase Identification Using Multitask Learning: Application to Hydrothermal Fluids

Traditional analysis of highly distorted micro X‐ray diffraction (μ‐XRD) patterns from hydrothermal fluid environments is a time‐consuming process, often requiring substantial data preprocessing and labeled experimental data. Herein, the potential of deep learning with a multitask learning (MTL) architecture to overcome these limitations is demonstrated. MTL models are trained to identify phase information in μ‐XRD patterns, minimizing the need for labeled experimental data and masking preprocessing steps. Notably, MTL models show superior accuracy compared to binary classification convolutional neural networks. Additionally, introducing a tailored cross‐entropy loss function improves MTL model performance. Most significantly, MTL models tuned to analyze raw and unmasked XRD patterns achieve close performance to models analyzing preprocessed data, with minimal accuracy differences. This work indicates that advanced deep learning architectures like MTL can automate arduous data handling tasks, streamline the analysis of distorted XRD patterns, and reduce the reliance on labor‐intensive experimental datasets.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Root Pulling Force Across Drought in Maize Reveals Genotype by Environment Interactions and Candidate Genes

High-throughput, field-based characterization of root systems for hundreds of genotypes in thousands of plots is necessary for breeding and identifying loci underlying variation in root traits and their plasticity. We designed a large-scale sampling of root pulling force, the vertical force required to extract the root system from the soil, in a maize diversity panel under differing irrigation levels for two growing seasons. We then characterized the root system architecture of the extracted root crowns. We found consistent patterns of phenotypic plasticity for root pulling force for a subset of genotypes under differential irrigation, suggesting that root plasticity is predictable. Using genome-wide association analysis, we identified 54 SNPs as statistically significant for six independent root pulling force measurements across two irrigation levels and four developmental timepoints. For every significant GWAS SNP for any trait in any treatment and timepoint we conducted post hoc tests for genotype-by-environment interaction, using a mixed model ANOVA. We found that 8 of the 54 SNPs showed significant GxE. Candidate genes underlying variation in root pulling force included those involved in nutrient transport. Although they are often treated separately, variation in the ability of plant roots to sense and respond to variation in environmental resources including water and nutrients may be linked by the genes and pathways underlying this variation. While functional validation of the identified genes is needed, our results expand the current knowledge of root phenotypic plasticity at the whole plant and gene levels, and further elucidate the complex genetic architecture of maize root systems.

Woods, Patrick↗

DOE OSTI2024

Exploring 2D X-ray diffraction phase fraction analysis with convolutional neural networks: Insights from kinematic-diffraction simulations

Abstract Deep-learning models are effective for analyzing the complex information in 2D X-ray diffraction (XRD) patterns. Accurately collecting parameters of the material sample is crucial during model training, significantly impacting model performance. In this study, we employ a kinematic-diffraction simulator to generate simulated 2D XRD patterns for Ti–6Al–4V alloy, allowing precise control of sample parameters. These simulated patterns are used to train convolutional neural networks, predicting $$\upbeta$$ β -phase volume fractions. The training data set consists exclusively of 2D XRD patterns with pure $$\upalpha$$ α - or pure $$\upbeta$$ β -phase, while the testing set incorporates patterns with intermediate phase volume fraction. In particular, we investigate how the architectures of the model influence prediction reliability and computational performance. Experimental results reveal that, with appropriate training, the convolutional neural network accurately detects intermediate phase volume fractions even trained with only pure-phase patterns, achieving a mean square error accuracy of $$9.4 \times 10^{-4}$$ 9.4 × 10 - 4 . Graphical abstract

Yue, Weiqi↗

DOE OSTI2021

Modelica-based system modeling for studying control-related faults in chiller plants and boiler plants serving large office buildings

System modeling is critical when studying operation faults in chiller plants and boiler plants. However, current fault models have difficulties faithfully representing the operation of chiller plants and boiler plants under the effects of those faults, especially the control-related ones. In this study, we present a systematic method to develop high-fidelity models for approximating the behaviors of chiller plants and boiler plants under faulty conditions. Compared to existing ones, the resulting fault models have two advantages: first, they better characterize the dynamic patterns in the system operation. In those models, control architecture and control logic are faithfully implemented. Thus, they can be used to study control-related faults, such as incorrect staging control due to sensor bias and mistuned feedback control. Second, they are readily extensible and can support large-scale investigations to explore different faulty conditions/scenarios. Those models are established in a hierarchical structure while modules in each layer can be redeclared and parameterized at upper layers. In such a case, modifications to the models can be realized through model modifiers and the process can be easily streamlined with scripts. We applied the proposed models in a comprehensive fault impact evaluation of the 13 control-related faults of chiller and boiler plants. In this evaluation, the proposed plant model is coupled with the EnergyPlus thermal load model to study the impact of various faulty scenarios. Based on the evaluation results, we identified the faults that have the most significant impacts on the operation of the chiller and boiler plants, respectively. We also found that the relationship between the impacts of the studied faults and the severity level of the faults are highly non-linear

Huang, Sen↗

DOE OSTI2025

Three-Dimensional Bubble Fluidics in Architected Porous Media

Gas bubble flows in porous media often exhibit complex and seemingly unpredictable behaviors that are difficult to control. This lack of control limits the ability to design effective devices which manage multiphase flows. Here, we show how the design of 3D printed pores can deterministically control the flow path of an injected gas stream. Open cell structures can be designed to shape the gas/liquid interface with fidelity to control how the two phases are distributed throughout a porous material. The distributed gas volume is free to interact physically and chemically with the surrounding liquid phase, an effect we exploit to create a logical control gate to redirect flows within a lattice. This also allows us to design architectures for reactive capture and aerating bioreactors, resulting in patterned boundaries which can make more effective use of the liquid and gas reagents.

3D microfluidics↗