Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “task work”

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 1,423 records · Page 79
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Performance Analysis of OpenSHMEM Applications with TAU Commander

The TAU Performance System® (TAU) is a powerful and highly versatile profiling and tracing tool ecosystem for performance engineering of parallel programs. Developed over the last twenty years, TAU has evolved with each new generation of HPC systems and scales efficiently to hundreds of thousands of cores. TAU’s organic growth has resulted in a loosely coupled software toolbox such that novice users first encountering TAU’s complexity and vast array of features are often intimidated and easily frustrated. To lower the barrier to entry for novice TAU users, ParaTools and the US Department of Energy have developed “TAU Commander,” a performance engineering workflow manager that facilitates a systematic approach to performance engineering, guides users through common profiling and tracing workflows, and offers constructive feedback in case of error. This work compares TAU and TAU Commander workflows for common performance engineering tasks in OpenSHMEM applications and demonstrates workflows targeting two different SHMEM implementations, Intel Xeon “Haswell” and “Knights Landing” processors, direct and indirect measurement methods, callsite, profiles, and traces.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Scaling up nanoscale water-driven energy conversion into evaporation-driven engines and generators

Evaporation is a ubiquitous phenomenon in the natural environment and a dominant form of energy transfer in the Earth’s climate. Engineered systems rarely, if ever, use evaporation as a source of energy, despite myriad examples of such adaptations in the biological world. In this work, we report evaporation-driven engines that can power common tasks like locomotion and electricity generation. These engines start and run autonomously when placed at air–water interfaces. They generate rotary and piston-like linear motion using specially designed, biologically based artificial muscles responsive to moisture fluctuations. Using these engines, we demonstrate an electricity generator that rests on water while harvesting its evaporation to power a light source, and a miniature car (weighing 0.1 kg) that moves forward as the water in the car evaporates. Evaporation-driven engines may find applications in powering robotic systems, sensors, devices and machinery that function in the natural environment.

42 ENGINEERING↗
DOE OSTI2014

Rationalizing the role of structural motif and underlying electronic structure in the finite temperature behavior of atomic clusters

Melting in finite size systems is an interesting but complex phenomenon. Many factors affect melting and owing to their interdependencies it is a challenging task to rationalize their roles in the phase transition. In this work, we demonstrate how structural motif of the ground state influences melting transition in small clusters. Here, we report a case with clusters of aluminum and gallium having same number of atoms, valence electrons, and similar structural motif of the ground state but drastically different melting temperatures. We have employed Born-Oppenheimer molecular dynamics to simulate the solid-like to liquid-like transition in these clusters. Our simulations have reproduced the experimental trends fairly well. Further, the detailed analysis of isomers has brought out the role of the ground state structure and underlying electronic structure in the finite temperature behavior of these clusters. For both clusters, isomers accessible before cluster melts have striking similarities and does have strong influence of the structural motif of the ground state. Further, the shape of the heat capacity curve is similar in both the cases but the transition is more spread over for Al{sub 36} which is consistent with the observed isomerization pattern. Our simulations also suggest a way to characterize transition region on the basis of accessibility of the ground state at a specific temperature.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Solvent-free synthesis of organometallic halides CH 3 NH 3 PbI 3 and (CH 3 NH 3 ) 3 Bi 2 I 9 and their thermoelectric transport properties

Organometallic halides are great candidates for optoelectronics. As an important family of semiconductors, understanding their thermoelectric transport properties is also important. This has been a challenging task as many of such compounds are highly intrinsic. In this work, we synthesized two halides, CH 3 NH 3 PbI 3 and (CH 3 NH 3 ) 3 Bi 2 I 9 , using a solvent-free method. We found an extraordinarily high Seebeck coefficient of +2600 ± 200 μ V/K in (CH 3 NH 3 ) 3 Bi 2 I 9 . For CH 3 NH 3 PbI 3 , our synthesis method led to a negative Seebeck coefficient of –1350 ± 50 μ V/K, in contrast to positive values observed in solvent synthesized samples. We also found the thermal conductivity of CH 3 NH 3 PbI 3 to be 0.38 W/m K, largely independent of temperature from 300 K to 450 K, despite a tetragonal-to-cubic phase transition. (CH 3 NH 3 ) 3 Bi 2 I 9 has an even lower thermal conductivity of 0.21 W/m K (also temperature independent) which is due to its soft phonon dispersion and weak bonds.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Super resolution convolutional neural network for feature extraction in spectroscopic data

Two dimensional (2D) peak finding is a common practice in data analysis for physics experiments, which is typically achieved by computing the local derivatives. However, this method is inherently unstable when the local landscape is complicated or the signal-to-noise ratio of the data is low. In this work, we propose a new method in which the peak tracking task is formalized as an inverse problem, which thus can be solved with a convolutional neural network (CNN). In addition, we show that the underlying physics principle of the experiments can be used to generate the training data. By generalizing the trained neural network on real experimental data, we show that the CNN method can achieve comparable or better results than traditional derivative based methods. This approach can be further generalized in different physics experiments when the physical process is known.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2020

Vectorization techniques for probability distribution functions using VecCore

Probability distribution functions (PDFs) are very used in modeling random processes and physics simulations. Improving the performance of algorithms that generate many random numbers under complex PDFs is often a very challenging task when methods as direct functions are not available. In this work we present general strategies on how to vectorize some PDFs using VecCore library. We show the results for the Exponential, Gaussian, discrete Poisson and Gamma probability distributions.

Chaparro Amaro, Oscar R.↗
DOE OSTI2022

Unsupervised quantum circuit learning in high energy physics

Unsupervised training of generative models is a machine learning task that has many applications in scientific computing. Here, in this work, we evaluate the efficacy of using quantum circuit-based generative models to generate synthetic data of high energy physics processes. We use nonadversarial, gradient-based training of quantum circuit Born machines to generate joint distributions over two and three variables.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Scaling Resolution of Gigapixel Whole Slide Images Using Spatial Decomposition on Convolutional Neural Networks

Gigapixel images are prevalent in scientific domains ranging from remote sensing, and satellite imagery to microscopy, etc. However, training a deep learning model at the natural resolution of those images has been a challenge in terms of both, overcoming the resource limit (e.g. HBM memory constraints), as well as scaling up to a large number of GPUs. In this paper, we trained Residual neural Networks (ResNet) on 22,528 x 22,528-pixel size images using a distributed spatial decomposition method on 2,304 GPUs on the Summit Supercomputer. We applied our method on a Whole Slide Imaging (WSI) dataset from The Cancer Genome Atlas (TCGA) database. WSI images can be in the size of 100,000 x 100,000 pixels or even larger, and in this work we studied the effect of image resolution on a classification task, while achieving state-of-the-art AUC scores. Moreover, our approach doesn't need pixel-level labels, since we're avoiding patching from the WSI images completely, while adding the capability of training arbitrary large-size images. This is achieved through a distributed spatial decomposition method, by leveraging the non-block fat-tree interconnect network of the Summit architecture, which enabled GPU-to-GPU direct communication. Finally, detailed performance analysis results are shown, as well as a comparison with a data-parallel approach when possible.

Tsaris, Aristeidis (aris)↗
DOE OSTI2025

Architecture-Aware Models of AI Engines for High-Performance Matrix Matrix Multiplication

The AI Engine (AIE) architecture, available in systems from mobile SoCs to server-class FPGAs, aims to efficiently execute AI/ML tasks through a two-dimensional array of compute tiles. Previous work on AIEs has explored different approaches to mapping computation across spatial arrays, but the compute kernel running on each tile has not been the focus. Additionally, the AIE-ML architecture introduces memory tiles and omits programmable logic, requiring new approaches to staging and moving data throughout the array. In this work we update analytical models developed for CPUs to produce the design of high performance kernels while introducing new model considerations such as memory structure, throughput, and latency as required by the AIE hardware. We evaluate our models by developing AIE-ML kernels for matrix multiplication in low-precision data types showing performance up to 95% of compute peak for the kernel when data resides in local memory and above 90% of compute peak when data resides in main memory.

Binder, Elliott D. [Carnegie Mellon University, Pi↗
DOE OSTI2020

TaskWorks v1.0

TaskWorks is a portable, high-level, task engine designed for HPC workloads. Applications can create tasks and define dependencies between them with the task engine. Once the task is defined and submitted to TaskWorks, the TaskWorks engine will execute it according to the specified dependencies, without additional input from the application. TaskWorks has an integrated event manager monitors files, sockets, timers, and MPI operations, and these events can be used as task dependencies. TaskWorks is compatible with MPI and is designed to work efficiently with HPC applications that perform.

Koziol, Quincey↗
DOE OSTI2025

Replacing non-biomedical concepts improves embedding of biomedical concepts

Embeddings are semantically meaningful representations of words in a vector space, commonly used to enhance downstream machine learning applications. Traditional biomedical embedding techniques often replace all synonymous words representing biological or medical concepts with a unique token, ensuring consistent representation and improving embedding quality. However, the potential impact of replacing non-biomedical concept synonyms has received less attention. Embedding approaches often employ concept replacement to replace concepts that span multiple words, such as non-small-cell lung carcinoma, with a single concept identifier (e.g., D002289). Also, all synonyms of each concept are merged into the same identifier. Here, we additionally leveraged WordNet to identify and replace sets of non-biomedical synonyms with their most common representatives. This combined approach aimed to reduce embedding noise from non-biomedical terms while preserving the integrity of biomedical concept representations. We applied this method to 1,055 biomedical concept sets representing molecular signatures or medical categories and assessed the mean pairwise distance of embeddings with and without non-biomedical synonym replacement. A smaller mean pairwise distance was interpreted as greater intra-cluster coherence and higher embedding quality. Embeddings were generated using the Word2Vec algorithm applied to a corpus of 10 million PubMed abstracts. Our results demonstrate that the addition of non-biomedical synonym replacement reduced the mean intra-cluster distance by an average of 8%, suggesting that this complementary approach enhances embedding quality. Future work will assess its applicability to other embedding techniques and downstream tasks. Python code implementing this method is provided under an open-source license.

algorithms↗
DOE OSTI2021

The mechanical response of micron-sized molecular crystals

Abstract Microstructures and corresponding properties of compacted powders ultimately depend on the mechanical response of individual particles. In principle, computational simulations can predict the results of powder compaction processes, but the selection of appropriate models for both particle–particle interactions and particle deformations across all relevant length scales remain nontrivial tasks, especially in material systems lacking detailed mechanical property information. The work presented here addresses these issues by conducting uniaxial compressions in situ inside of a scanning electron microscope to characterize the mechanical response of individual micron-sized particles of a molecular crystal, hexanitrohexaazaisowurtzitane (CL-20). This experimental approach enabled the collection of quantitative force and displacement data alongside simultaneous imaging to capture morphology changes. The results reveal information about elastic deformation, yield, plastic deformation, creep, and fracture phenomena. Accordingly, this work demonstrates a generalizable approach for assessing the mechanical response of individual micron-sized molecular crystal particles and utilizing those responses in particle-level models. Graphic abstract

36 MATERIALS SCIENCE↗
DOE OSTI2013

Final Report - DE-EE0000260

In this program, we are developing next-generation SOFC materials and architectures for use with biomass-derived fuel streams, with a focus on addressing the technical challenges and the required operating windows associated with carbon-deposit formation. We are developing fuel-processing strategies for processing landfill- and anaerobic-digester-derived biomass streams for use in SOFCs. Similarly, we are examining processing of biomass-derived liquid fuels for use in solid-oxide fuel cell systems. We conduct extensive computational modeling work to support the materials-development and fuel-processing tasks. Modeling is conducted across a broad range of length scales, from the microscopic level to examine the fundamental underpinnings of cell operation, to the mesoscopic level to examine and optimize fuel-reformer design, and to the system level to craft optimal integration and dynamic-control solutions.

03 NATURAL GAS↗
DOE OSTI2016

Streaming Visual Analytics Workshop Report

How can we best enable users to understand complex emerging events and make appropriate assessments from streaming data? This was the central question addressed at a three-day workshop on streaming visual analytics. This workshop was organized by Pacific Northwest National Laboratory for a government sponsor. It brought together forty researchers and subject matter experts from government, industry, and academia. This report summarizes the outcomes from that workshop. It describes elements of the vision for a streaming visual analytic environment and set of important research directions needed to achieve this vision. Streaming data analysis is in many ways the analysis and understanding of change. However, current visual analytics systems usually focus on static data collections, meaning that dynamically changing conditions are not appropriately addressed. The envisioned mixed-initiative streaming visual analytics environment creates a collaboration between the analyst and the system to support the analysis process. It raises the level of discourse from low-level data records to higher-level concepts. The system supports the analyst’s rapid orientation and reorientation as situations change. It provides an environment to support the analyst’s critical thinking. It infers tasks and interests based on the analyst’s interactions. The system works as both an assistant and a devil’s advocate, finding relevant data and alerts as well as considering alternative hypotheses. Finally, the system supports sharing of findings with others. Making such an environment a reality requires research in several areas. The workshop discussions focused on four broad areas: support for critical thinking, visual representation of change, mixed-initiative analysis, and the use of narratives for analysis and communication.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2018

Investigation of Coupled Processes in Argillite Rock: FY18 Progress

The report presents the current status of the TOUGH-FLAC and TOUGH-RBSN simulators for modeling of coupled THM processes in argillite, including fracturing, and verification of the TOUGH-FLAC model against major in situ heater experiments at the Mont Terri Underground Research Laboratory (URL) in Switzerland and at the URL in Bure, France. The heater experiments modeled are the Mont Terri FE (Full-scale Emplacement) Experiment, conducted as part of the Mont Terri Project, and the TED and ALC experiments conducted in Callovo-Oxfordian claystone (COx) at the Meuse/Haute-Marne (MHM) URL in Bure, France. The modeling of the TED and ALC heater experiments is part of a modeling task (Task E) of the international DECOVALEX-2019 project. DECOVALEX, which stands for DEvelopment of COupled Models and their VALidation against EXperiments, is an international collaborative activity, in which DOE and LBNL gain access to unique laboratory and field data defined as modeling test cases. DECOVALEX tasks are studied collectively among several international groups to better understand the processes and to improve numerical models that could eventually be applied in the performance assessment for nuclear waste disposal in clay host rocks and bentonite backfill. The report also includes a description of LBNL’s activities for modeling gas migration in bentonite related to Task A of the international DECOVALEX project, and presents new FY2018 work related to monitoring and characterization of faults using a unique borehole tool.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2017

Europa Lander Modeling Transport Capabilities Survey

The following trade study was done to answer the following task from the Sandia JPL Collaboration for Europa Lander Statement of Work: Survey SNL capabilities for modeling the transport and survivability of biological organisms in extremely hot, cold, and high radiation environments.

63 RADIATION, THERMAL, AND OTHER ENVIRON. POLLUTAN↗
DOE OSTI2021

Europa Lander Aseptic Integration Facilities Survey

The following trade study was done to answer the following task from the Sandia JPL Collaboration for Europa Lander Statement of Work: Survey facility infrastructure SNL may have for performing aseptic assembly and integration of S/C and assess its suitability for PP applications.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

[Europa Lander] Decontamination Methods Trade Study

The following trade study was done to answer the following task from the Sandia JPL Collaboration for Europa Lander Statement of Work: Perform a trade study to assess the feasibility of other sterilization/decontamination methods for reducing forward biological contamination on S/C and assess their suitability for PP applications

42 ENGINEERING↗