Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Amazon”

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 271 records · Page 15
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

MassiveGNN: Efficient Training via Prefetching for Massively Connected Distributed Graphs

Graph Neural Networks (GNN) are indispensable in learning from graph-structured data, yet their rising computational costs, especially on massively connected graphs, pose significant challenges in terms of execution performance. To tackle this, distributed-memory solutions such as partitioning the graph to concurrently train multiple replicas of GNNs are in practice. However, approaches requiring a partitioned graph usually suffer from communication overhead and load imbalance, even under optimal partitioning and communication strategies due to irregularities in the neighborhood minibatch sampling. This paper proposes practical trade-offs for improving the sampling and communication overheads for representation learn- ing on distributed graphs (using popular GraphSAGE architecture) by developing a parameterized prefetch and eviction scheme on top of the state-of-the-art Amazon DistDGL distributed GNN framework, demonstrating about 15–40% improvement in end-to-end training performance on the NERSC Perlmutter supercomputer for various OGB datasets.

Machine Leanring, high performance comptuing, grap↗
DOE OSTI2022

Investigating Users’ Privacy Concerns of Internet of Things (IoT) Smart Devices

Although the number of smart Internet of Things (IoT) devices has grown in recent years, the public's perception of how effectively these devices secure IoT data has been questioned. Many IoT users do not have a good level of confidence in the security or privacy procedures implemented within IoT smart devices for protecting personal IoT data. Moreover, determining the level of confidence end users have in their smart devices is becoming a major challenge. In this paper, we present a study that focuses on identifying privacy concerns IoT end users have when using IoT smart devices. We investigated multiple smart devices and conducted a survey to identify users’ privacy concerns. Furthermore, we identify five IoT privacy-preserving (IoTPP) control policies that we define and employ in comparing the privacy measures implemented by various popular smart devices. Results from our study show that the over 86% of participants are very or extremely concerned about the security and privacy of their personal data when using smart IoT devices such as Google Nest Hub or Amazon Alexa. In addition, our study shows that a significant number of IoT users may not be aware that their personal data is collected, stored or shared by IoT devices.

Joy, Daniel↗
DOE OSTI2024

Escape Ratio Contributes More Than Fluorescence Yield to SIF-GPP Relationship Over Crops and Rainforest

Solar-induced chlorophyll fluorescence (SIF) is an effective indicator to track the gross primary productivity (GPP). However, there is still a lack of a clear understanding for the contribution of physiological and structural factors to the SIF–GPP relationship at the canopy scale. To quantify the influence of different SIF components, particularly the photon escape ratio ($f$ esc ) and fluorescence yield (Φ F ), on the SIF–GPP relationship, this study evaluated the performance of various vegetation indices (VIs), SIF components, light use efficiency (LUE), and GPP over two typical biomes (crops and rainforest), using a range of satellite remote sensing products. In August of each year from 2018 to 2020, both SIF and GPP over United States (U.S.) Corn Belt are higher than those over the Amazon rainforest, attributed to the consistent pattern of higher $f$ esc and LUE over crops than over rainforest, as well as Φ F . Furthermore, the structural signals represented by $f$ esc (R = 0.41–0.64) can better capture the LUE variations than Φ F (R = 0.10–0.30) for each biome. In conclusion, this study highlights that $f$ esc , determined by canopy structure, has great potential to capture LUE and GPP changes within and across biomes.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Secure Federated Learning Across Heterogeneous Cloud and High-Performance Computing Resources: A Case Study on Federated Fine-Tuning of LLaMA 2

Federated learning enables multiple data owners to collaboratively train robust machine learning models without transferring large or sensitive local datasets by only sharing the parameters of the locally trained models. Here, in this article, we elaborate on the design of our Advanced Privacy-Preserving Federated Learning (APPFL) framework, which streamlines end-to-end secure and reliable federated learning experiments across cloud computing facilities and high-performance computing resources by leveraging Globus Compute, a distributed function as a service platform, and Amazon Web Services. We further demonstrate the use case of APPFL in fine-tuning an LLaMA 2 7B model using several cloud resources and supercomputers.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Stimulation of isoprene emissions and electron transport rates are a key mechanism of thermal tolerance in the tropical species Vismia guianensis

Tropical forests absorb large amounts of atmospheric CO 2 through photosynthesis, but high surface temperatures suppress this absorption while promoting isoprene emissions. While mechanistic isoprene emission models predict a tight coupling to photosynthetic electron transport (ETR) as a function of temperature, direct field observations of these phenomenon are lacking in the tropics and are necessary to assess the impact of a warming climate on global isoprene emissions. In this paper, we demonstrate that in the early successional species Vismia guianensis in the central Amazon, ETR rates increased with temperature in concert with isoprene emissions, even as stomatal conductance ($g _s$) and net photosynthetic carbon fixation ($P_n$) declined. We observed the highest temperatures of continually increasing isoprene emissions yet reported (50°C). While $P_n$ showed an optimum value of 32.6 ± 0.4°C, isoprene emissions, ETR, and the oxidation state of PSII reaction centers (q L ) increased with leaf temperature with strong linear correlations for ETR (ρ = 0.98) and q L (ρ= 0.99) with leaf isoprene emissions. In contrast, other photoprotective mechanisms, such as non-photochemical quenching (NPQ), were not activated at elevated temperatures. Inhibition of isoprenoid biosynthesis repressed $P_n$ at high temperatures through a mechanism that was independent of stomatal closure. While extreme warming will decrease $g_s$ and $P_n$ in tropical species, our observations support a thermal tolerance mechanism where the maintenance of high photosynthetic capacity under extreme warming is assisted by the simultaneous stimulation of ETR and metabolic pathways that consume the direct products of ETR including photorespiration and the biosynthesis of thermoprotective isoprenoids. Our results confirm that models which link isoprene emissions to the rate of ETR hold true in tropical species and provide necessary "ground-truthing" for simulations of the large predicted increases in tropical isoprene emissions with climate warming.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2021

Understanding water and energy fluxes in the Amazonia: Lessons from an observation‐model intercomparison

Abstract Tropical forests are an important part of global water and energy cycles, but the mechanisms that drive seasonality of their land‐atmosphere exchanges have proven challenging to capture in models. Here, we (1) report the seasonality of fluxes of latent heat (LE), sensible heat ( H ), and outgoing short and longwave radiation at four diverse tropical forest sites across Amazonia—along the equator from the Caxiuanã and Tapajós National Forests in the eastern Amazon to a forest near Manaus, and from the equatorial zone to the southern forest in Reserva Jaru; (2) investigate how vegetation and climate influence these fluxes; and (3) evaluate land surface model performance by comparing simulations to observations. We found that previously identified failure of models to capture observed dry‐season increases in evapotranspiration (ET) was associated with model overestimations of (1) magnitude and seasonality of Bowen ratios (relative to aseasonal observations in which sensible was only 20%–30% of the latent heat flux) indicating model exaggerated water limitation, (2) canopy emissivity and reflectance (albedo was only 10%–15% of incoming solar radiation, compared to 0.15%–0.22% simulated), and (3) vegetation temperatures (due to underestimation of dry‐season ET and associated cooling). These partially compensating model‐observation discrepancies (e.g., higher temperatures expected from excess Bowen ratios were partially ameliorated by brighter leaves and more interception/evaporation) significantly biased seasonal model estimates of net radiation ( R n ), the key driver of water and energy fluxes (LE ~ 0.6 R n and H ~ 0.15 R n ), though these biases varied among sites and models. A better representation of energy‐related parameters associated with dynamic phenology (e.g., leaf optical properties, canopy interception, and skin temperature) could improve simulations and benchmarking of current vegetation–atmosphere exchange and reduce uncertainty of regional and global biogeochemical models.

Restrepo‐Coupe, Natalia↗
DOE OSTI2026

Habitat Affinity of Riverine Dissolved Organic Matter Linked to Molecular Traits

Fluxes of organic matter across Earth habitats shape ecosystem function and carbon storage. Yet, it is challenging to predict howthe molecules are preferentially accumulated in specific habitats, that is, the habitat affinities, due to their diverse character-istics. Here, we develop an indicator of compositional-level habitat affinity for dissolved organic matter (DOM) by quantifyingthe aggregated affinities of individual molecules that are over- or under-represented relative to a reference habitat. Applyingthis indicator to 93 paired riverine water-sediment sites, we find that 65.1% of molecules show non-significant habitat affinitiesfor water or sediment and are thermodynamically favorable and susceptible to microbial degradation. Consistently, the habitataffinities of DOM assemblages decrease with lower molecular weights and recalcitrance for waters, and with higher thermody-namic favorability for sediments. These affinities are both reduced by elevated temperatures and anthropogenic pressures, suchas nitrogen loading and land-use change. Our projection of habitat affinities of global riverine DOM shows that regions withintense anthropogenic disturbances or warm climates, such as East Asia, Western Europe, and the Amazon, tend to exhibit loweraffinities. These findings suggest enhanced molecular homogenization between water and sediment habitats is associated withenvironmental changes and underscore the importance of habitat affinities for predicting the potential trajectories of organiccarbon across ecosystems

Cui, Yifan↗
DOE OSTI2021

SQL and NoSQL Databases for Cyber Physical Production Systems in Internet of Things for Manufacturing (IoTfM)

Abstract In this paper, the design and performance differences between Relational Database Management Systems (RDBMS) and NoSQL Database Systems are examined, with attention to their applicability for real-world Internet of Things for manufacturing (IoTfM) data. While previous work has extensively compared SQL and NoSQL for both generalized and IoT uses, this work specifically examines the tradeoffs and performance differences for manufacturing applications by using a high-fidelity data set collected from a large US manufacturing firm. Growing an IoT system beyond the pilot stage requires scalable data storage; this work seeks to determine the impact of selected database systems on data write performance at scale. Payload size and message frequency were used as the primary characteristics to maintain model fidelity in simulated clients. As the number of simulated asset clients grow, the data write latency was calculated to determine how both database systems’ performance were affected. To isolate the RDBMS and NoSQL differences, a cloud environment was created using Amazon Web Services (AWS) with two identical data ingestion pipelines: writing data to an RDMBS (1) using AWS Aurora MySQL, and (2) using AWS DynamoDB NoSQL. The findings may provide guidance for further experimentation in large-scale manufacturing IoT implementations.

Gamero, David↗
DOE OSTI2020

Atmospheric fungal nanoparticle bursts

Aerosol nanoparticles play an important role in the climate system by affecting cloud formation and properties, as well as in human health because of their deep reach into lungs and the circulatory system. Determining nanoparticle sources and composition is a major challenge in assessing their impacts in these areas. The sudden appearance of large numbers of atmospheric nanoparticles is commonly attributed to secondary formation from gas-phase precursors, but in many cases, the evidence for this is equivocal. We report the detection of a mode of fungal fragments with a mobility diameter of roughly 30 nm released in episodic bursts in ambient air over an agricultural area in northern Oklahoma. These events reached concentrations orders of magnitude higher than other reports of biological particles and show similarities to unclarified events reported previously in the Amazon. These particles potentially represent a large source of both cloud-forming ice nuclei and respirable allergens in a variety of ecosystems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Compiler-Driven FPGA Virtualization with SYNERGY

FPGAs are increasingly common in modern applications, and cloud providers now support on-demand FPGA acceleration in datacenters. Applications in datacenters run on virtual infrastructure, where consolidation, multi-tenancy, and workload migration enable economies of scale that are fundamental to the provider's business. However, a general strategy for virtualizing FPGAs has yet to emerge. While manufacturers struggle with hardware-based approaches, we propose a compiler/runtime-based solution called Synergy. We show a compiler transformation for Verilog programs that produces code able to yield control to software atsub-clock-tickgranularity according to the semantics of the original program. Synergy uses this property to efficiently support core virtualization primitives: suspend and resume, program migration, and spatial/temporal multiplexing, on hardware which is availabletoday.We use Synergy to virtualize FPGA workloads across a cluster of Intel SoCs and Xilinx FPGAs on Amazon F1. The workloads require no modification, run within 3--4x of unvirtualized performance, and incur a modest increase in FPGA fabric usage.

Computer Science↗
DOE OSTI2025

A Digital Twin of Scalable Quantum Clouds

Quantum computing has emerged as a transformative technology capable of solving complex problems beyond the limit of classical systems. The rapid development of quantum processors has led to the proliferation of cloud-based quantum computing services offered by platforms such as IBM, Google, and Amazon. These platforms introduce unique challenges in resource allocation, job scheduling, and multi-device orchestration as quantum workloads become increasingly complex. In this work, we present a digital twin of quantum cloud infrastructures: a framework designed to model and simulate the behavior of real quantum cloud systems. Developed in Python using the SimPy discrete-event simulation library, the framework replicates key aspects of quantum cloud environments, including detailed quantum device modeling, job lifecycle management, and job fidelity. It incorporates noise-aware fidelity estimation, making it the first of its kind to simulate superconducting gate-based quantum cloud systems at an administrative level with job fidelity. We present use cases as proof of concept, demonstrating that our quantum cloud simulation framework can act as a digital twin of a quantum cloud and support the modeling and implementation of practical systems.

Luo, Waylon [Kent State University]↗
DOE OSTI2020

materialsCloudCompute

materialsCloudCompute (mCC) is a software package for running easily parallelizable materials science theoretical computations on Amazon Web Services (AWS) infrastructure. mCC provides tools to partition out computational problem into smaller subproblems which can be run in parallel, and ensure that each of those subproblems completes successfully. It utilizes the following AWS services: EC2, S3, and RDS. It operates by launch a managing EC2 instance, which can launch worker instances that perform the calculations. The Managing instance tracks the subproblems and ensures that the Worker instances continue to operation, killing and relaunching them if they fail.

Forbes, David↗
DOE OSTI2022

s3-browser

Web application that interacts with the Amazon AWS S3 REST API to allow users to manage files and buckets without having to use the aws command line application.

Chu, Vincent↗
DOE OSTI2021

Learning Universal Authorship Representations

This code contains all the utilities required to reproduce the results of our EMNLP 2021 paper "Learning Universal Authorship Representations". It contains the utilities required for download the datasets, training our model, and performing all evaluations necessary for reproducing the results in the paper. Here's the abstract of our work: Determining whether two documents were composed by the same author, also known as authorship verification, has traditionally been tackled using statistical methods. Recently, authorship representations learned using neural networks have been found to outperform alternatives, particularly in large-scale settings involving hundreds of thousands of authors. But do such representations learned in a particular domain transfer to other domains? Or are these representations inherently entangled with domain-specific features? To study these questions, we conduct the first large-scale study of cross-domain transfer for authorship verification considering zero-shot transfers involving three disparate domains: Amazon reviews, fanfiction short stories, and Reddit comments. We find that although a surprising degree of transfer is possible between certain do- mains, it is not so successful between others. We examine properties of these domains that influence generalization and propose simple but effective methods to improve transfer.

Rivera Soto, RafaelA↗
DOE OSTI2023

Azure Data Tools (FKA: OEDI (Open Energy Data Initiative) Data Access Tools) [SWR-23-92]

The Open Energy Data Initiative (OEDI) provides a number of tools to enable the use of the open data published through this initiative. The source is largely written in Python, including Jupyter notebooks. The Open Energy Data Initiative (OEDI) is a partnership between the National Renewable Energy Laboratory (NREL), the U.S. Department of Energy (DOE), Amazon, Microsoft, and Google to provide universal access to big data in the cloud. At the heart of OEDI is a centralized repository of high-value energy research datasets aggregated from DOE Program Offices, National Laboratories and other collaborators.

Jonathan, Weers↗
DOE OSTI2020

Pronounced Impact of Salinity on Rapidly Intensifying Tropical Cyclones

We show the importance of salinity for rapidly intensifying Atlantic tropical cyclones and demonstrate the potential for improved prediction of rapid intensification through the inclusion of salinity. Tropical Cyclone (TC) rapid intensification (RI) is difficult to predict and poses a formidable threat to coastal populations. A warm upper ocean is well-known to favor RI, but the role of ocean salinity is less clear. This study shows a strong inverse relationship between salinity and TC RI in the eastern Caribbean and western tropical Atlantic due to near-surface freshening from the Amazon-Orinoco River system. In this region, rapidly intensifying TCs induce a much stronger surface enthalpy flux compared to more weakly intensifying storms, in part due to a reduction in SST cooling caused by salinity stratification. This reduction has a noticeable positive impact on TCs undergoing RI, but the impact of salinity on more weakly intensifying storms is insignificant. These statistical results are confirmed through experiments with an ocean mixed layer model, which show that the salinity-induced reduction in SST cold wakes increases significantly as the storm’s intensification rate increases. Currently, operational statistical-dynamical RI models do not use salinity as a predictor. Through experiments with a statistical RI prediction scheme, it is found that the inclusion of surface salinity significantly improves the RI detection skill, offering promise for improved operational RI prediction. Satellite surface salinity may be valuable for this purpose, given its global coverage and availability in near real-time.

Balaguru, Karthik↗
DOE OSTI2022

Large-Eddy Simulations of Convection Initiation over Heterogeneous, Low Terrain

Abstract Large-eddy simulations are conducted to investigate and physically interpret the impacts of heterogeneous, low terrain on deep-convection initiation (CI). The simulations are based on a case of shallow-to-deep convective transition over the Amazon River basin, and use idealized terrains with varying levels of ruggedness. The terrain is designed by specifying its power-spectral shape in wavenumber space, inverting to physical space assuming random phases for all wave modes, and scaling the terrain to have a peak height of 200 m. For the case in question, these modest terrain fields expedite CI by up to 2–3 h, largely due to the impacts of the terrain on the size of, and subcloud support for, incipient cumuli. Terrain-induced circulations enhance subcloud kinetic energy on the mesoscale, which is realized as wider and longer-lived subcloud circulations. When the updraft branches of these circulations breach the level of free convection, they initiate wider and more persistent cumuli that subsequently undergo less entrainment-induced cloud dilution and detrainment-induced mass loss. As a result, the clouds become more vigorous and penetrate deeper into the troposphere. Larger-scale terrains are more effective than smaller-scale terrains in promoting CI because they induce larger enhancements in both the width and the persistence of subcloud updrafts.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Turbulent Kinetic Energy Budgets over Gentle Topography Covered by Forests

Large-eddy simulations of flow over a “horizontally” uniform model forest are used to investigate the effects of gentle topography on the turbulent kinetic energy (TKE) budget within the canopy roughness sublayer. Despite significant differences between simulations using idealized sinusoidal topography and real topography of the Amazon forest, results indicate that the effects of topography are located predominantly in the upper canopy and above, and are mostly caused by mean advection of TKE. The “horizontally” averaged TKE budget from idealized and real gentle topographies are almost identical to that for flat terrain, including a clear inertial layer above the roughness sublayer in which shear production is balanced by local dissipation. At topography crests, where observational towers are usually located, mean vertical advection of TKE can be as important as horizontal advection. Here, we propose the use on an approximate TKE balance equation to estimate mean advection from single tower measurements, and introduce a new advection index that can be used as a proxy to quantify the importance of the topography on the TKE budget.

54 ENVIRONMENTAL SCIENCES↗