Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “distributed processing”

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 847 records · Page 47

Distributed Storage Systems for Data Intensive Computing

In this chapter, the authors present an overview of the utility of distributed storage systems in supporting modern applications that are increasingly becoming data intensive. Their coverage of distributed storage systems is based on the requirements imposed by data intensive computing and not a mere summary of storage systems. To this end, they delve into several aspects of supporting data-intensive analysis, such as data staging, offloading, checkpointing, and end-user access to terabytes of data, and illustrate the use of novel techniques and methodologies for realizing distributed storage systems therein. The data deluge from scientific experiments, observations, and simulations is affecting all of the aforementioned day-to-day operations in data-intensive computing. Modern distributed storage systems employ techniques that can help improve application performance, alleviate I/O bandwidth bottleneck, mask failures, and improve data availability. They present key guiding principles involved in the construction of such storage systems, associated tradeoffs, design, and architecture, all with an eye toward addressing challenges of data-intensive scientific applications. They highlight the concepts involved using several case studies of state-of-the-art storage systems that are currently available in the data-intensive computing landscape.

99 GENERAL AND MISCELLANEOUS↗

File I/O for MPI Applications in Redundant Execution Scenarios

As multi-petascale and exa-scale high-performance computing (HPC) systems inevitably have to deal with a number of resilience challenges, such as a significant growth in component count and smaller circuit sizes with lower circuit voltages, redundancy may offer an acceptable level of resilience that traditional fault tolerance techniques, such as checkpoint/restart, do not. Although redundancy in HPC is quite controversial due to the associated cost for redundant components, the constantly increasing number of cores-per-processor is tilting this cost calculation toward a system design where computation, such as for redundancy, is much cheaper and communication, needed for checkpoint/restart, is much more expensive. Recent research and development activities in redundancy for Message Passing Interface (MPI) applications focused on availability/reliability models and replication algorithms. This paper takes a first step toward solving an open research problem associated with running a parallel application redundantly, which is file I/O under redundancy. The approach intercepts file I/O calls made by a redundant application to employ coordination protocols that execute file I/O operations in a redundancy-oblivious fashion when accessing a node-local file system, or in a redundancy-aware fashion when accessing a shared networked file system. A proof-of concept prototype is presented and a number of coordination protocols are described and evaluated. The results show the performance impact for redundantly accessing a shared networked file system, but also demonstrate the capability to regain performance by utilizing MPI communication between replicas and parallel file I/O.

99 GENERAL AND MISCELLANEOUS↗

SWAMP+: multiple subsequence alignment using associative massive parallelism

A new parallel algorithm SWAMP+ incorporates the Smith-Waterman sequence alignment on an associative parallel model known as ASC. It is a highly sensitive parallel approach that expands traditional pairwise sequence alignment. This is the first parallel algorithm to provide multiple non-overlapping, non-intersecting subsequence alignments with the accuracy of Smith-Waterman. The efficient algorithm provides multiple alignments similar to BLAST while creating a better workflow for the end users. The parallel portions of the code run in O(m+n) time using m processors. When m = n, the algorithmic analysis becomes O(n) with a coefficient of two, yielding a linear speedup. Implementation of the algorithm on the SIMD ClearSpeed CSX620 confirms this theoretical linear speedup with real timings.

99 GENERAL AND MISCELLANEOUS↗

Coherent x-ray diffraction imaging of paint pigmentparticles by scanning a phase plate modulator

We have implemented a coherent x-ray diffraction imaging technique that scans a phase plate to modulate wave-fronts of the x-ray beam transmitted by samples. The method was applied to measure a decorative alkyd paint containing iron oxide red pigment particles. By employing an iterative algorithm for wave-front modulation phase retrieval, we obtained an image of the paint sample that shows the distribution of the pigment particles and is consistent with the result obtained from a transmission x-ray microscope. The technique has been experimentally proven to be a feasible coherent x-ray imaging method with about 120 nm spatial resolution and was shown to work well with industrially relevant specimens.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Single- and Two-Photon-Induced Processes at the B Factories

We discuss single- and two-photon-induced processes in e{sup +}e{sup -} annihilations with center-of-mass energy near 10.58 GeV from the BaBar and Belle experiments. In particular, we present experimental results from two-photon physics of {gamma}{gamma} {yields} {pi}{sup 0}{pi}{sup 0} and {gamma}{gamma}* {yields} {pi}{sup 0}. We also review the observation of the Two-Virtual-Photon-Annihilation process (e{sup +}e{sup -} {yields} {rho}{sup 0}{rho}{sup 0} and e{sup +}e{sup -} {yields} {phi}{rho}{sup 0}) and the observation of e{sup +}e{sup -} {yields} {rho}{sup +}{rho}{sup -}, which should be primarily a one virtual photon process, but whose angular distributions may imply potential interference effects.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A Comparison of Techniques for Detecting Abnormal Change in Blogs

Distributed collections are made of metadata entries that contain references to artifacts not controlled by the collection curators. These collections often have limited forms of change; for digital distributed collections, primarily creation and deletion of additional resources. However, there exists a class of digital collection that undergoes additional kinds of change. These collections consist of resources that are distributed across the Internet and brought together via hyperlinking. Resources in these collections can be expected to change as time goes on. Part of the difficulty in maintaining these collections is determining if a changed page is still a valid member of the collection. Others have tried to address this by defining a maximum allowed threshold of change, however, these methods treat change as a potential problem and treat web content as static despite its intrinsic dynamicism. Instead we acknowledge change on the web as a normal part of a web document and determine the difference between what a maintainer expects a page to do and what it actually does. In this work we evaluate options for extractors and analyzers from a suite of techniques against a human-generated ground-truth set of blog changes. The results of this work show a statistically significant improvement over traditional threshold techniques for our collection.

99 GENERAL AND MISCELLANEOUS↗

Rehabilitation of notch damaged steel beam using a carbon fiber reinforced multiphase-matrix composite

The retrofit of notch damaged steel beams is investigated via the experimental testing of nine wide-flange steel beam specimens and finite element simulation. Three notch configurations representing various damage levels were identified, and the beam specimens were retrofitted using CFRP laminates and a recently developed polymeric matrix composite - CarbonFlex - that exhibits superior energy dissipation and ductility properties, where the peak-load deflections were between 49.4% and 65.2% higher using the CarbonFlex-retrofitted beams. The results are attributed to the substantially higher damage tolerance capability of CarbonFlex than conventional CFRP. Finite element models were developed to investigate the damage processes and strain/ stress distributions near the notch tips. The numerical results match closely with the experimentally determined load-deflection curves and the strain fields obtained by the digital imaging correlations (DIC) system. Both experimental and numerical results clearly indicate the effectiveness of CarbonFlex, as a candidate retrofitting material, to retrofit damaged steel structures. Lastly, the micro-mechanisms by which CarbonFlex could sufficiently sustain a significant amount of the peak strength at large displacement ductility values are discussed with the aid of scanning electron microscopy (SEM) pictures.

Zhou, HongYu↗

Real-time Process Monitoring and Temperature Mapping of the 3D Polymer Printing Process

An extended range IR camera was used to make temperature measurements of samples as they are being manufactured. The objective is to quantify the temperature variation inside the system as parts are being fabricated, as well as quantify the temperature of a part during fabrication. The IR camera was used to map the temperature within the build volume of the oven and surface temperature measurement of a part as it was being manufactured. The development of the temperature map of the oven provides insight into the global temperature variation within the oven that may lead to understanding variations in the properties of parts as a function of location. The observation of the temperature variation of a part that fails during construction provides insight into how the deposition process itself impacts temperature distribution within a single part leading to failure.

3D Printing↗

Alternate Endpoints for Deep Vadose Zone Environments: Challenges, Opportunities, and Progress

Current requirements for site remediation and closure are standards-based and are often overly conservative, costly, and in some cases, technically impractical to achieve. Use of risk-informed alternate endpoints provide a means to achieve remediation goals that are permitted by regulations and are protective of human health and the environment. Alternate endpoints enable establishing a path for cleanup that may include intermediate remedial milestones and transition points and/or regulatory alternatives to standards-based remediation. A framework is presented that is centered around developing and refining conceptual models in conjunction with assessing risks and potential endpoints as part of a system-based assessment that integrates site data with scientific understanding of processes that control the distribution and transport of contaminants in the subsurface and pathways to receptors. This system based assessment and subsequent implementation of the remediation strategy with appropriate monitoring are targeted at providing a holistic approach to addressing risks to human health and the environment. This holistic approach also enables effective predictive analysis of contaminant behavior to provide defensible criteria and data for making long-term decisions. Developing and implementing an alternate endpoint-based approach for remediation and waste site closure presents a number of challenges and opportunities. Categories of these challenges include scientific and technical, regulatory, institutional, and budget and resource allocation issues. Opportunities exist for developing and implementing systems-based approaches with respect to supportive characterization, monitoring, predictive modeling, and remediation approaches.

endpoints↗

Modular space vehicle boards, control software, reprogramming, and failure recovery

A space vehicle may have a modular board configuration that commonly uses some or all components and a common operating system for at least some of the boards. Each modular board may have its own dedicated processing, and processing loads may be distributed. The space vehicle may be reprogrammable, and may be launched without code that enables all functionality and/or components. Code errors may be detected and the space vehicle may be reset to a working code version to prevent system failure.

33 ADVANCED PROPULSION SYSTEMS↗

The SeaQuest Spectrometer at Fermilab

The SeaQuest spectrometer at Fermilab was designed to detect oppositely-charged pairs of muons (dimuons) produced by interactions between a 120 GeV proton beam and liquid hydrogen, liquid deuterium and solid nuclear targets. The primary physics program uses the Drell-Yan process to probe antiquark distributions in the target nucleon. The spectrometer consists of a target system, two dipole magnets and four detector stations. The upstream magnet is a closed-aperture solid iron magnet which also serves as the beam dump, while the second magnet is an open aperture magnet. Each of the detector stations consists of scintillator hodoscopes and a high-resolution tracking device. The FPGA-based trigger compares the hodoscope signals to a set of pre-programmed roads to determine if the event contains oppositely-signed, high-mass muon pairs.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Energy Resilience Through Grid Modernization and Renewables Integration

The National Renewable Energy Laboratory (NREL) is studying renewable energy system integration and security and resilience planning as part of the U.S. Department of Energy's Grid Modernization Initiative. Some recent work focused on incorporating the Value of Resilience into the decision making process for investing in distributed, behind-the-meter energy systems to meet critical loads during grid outages. Cost / benefit analyses show how accounting for the Value of Resilience as a benefit changes solar PV and battery energy storage system capacities for maximum owner benefit.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Modular space vehicle boards, control software, reprogramming, and failure recovery

A space vehicle may have a modular board configuration that commonly uses some or all components and a common operating system for at least some of the boards. Each modular board may have its own dedicated processing, and processing loads may be distributed. The space vehicle may be reprogrammable, and may be launched without code that enables all functionality and/or components. Code errors may be detected and the space vehicle may be reset to a working code version to prevent system failure.

97 MATHEMATICS AND COMPUTING↗

Real-time Intrusion Detection for High-bandwidth Research Networks using Unsupervised Deep Learning

Nowadays, data generated by large-scale scientific experiments are on the scale of petabytes per month. These data are transferred through dedicated high-bandwidth networks across distributed collaboration sites for processing, storage, and analysis. Like general purpose networks, research networks experience attacks. However, monitoring traffic in such high-bandwidth network environments is challenging with conventional deep packet inspection (DPI)- based network intrusion detection systems (NIDS). Recently, machine learning (ML) algorithms have shown potential in detecting anomalous patterns with better generalizability and faster speed. Compared to the general purpose networks, traffic movement in research network is less volatile and more likely to benefit from the ML-based anomaly analysis. We present ThunderSecure, an efficient AI-powered system for detecting anomalies in 100G research networks. ThunderSecure implements a high-throughput packet processing infrastructure using multi-cores and GPUs. It takes time-series statistics from a variety of network measurements and uses 2D Convolutional Neural Network (CNN) to explore both spatial and temporal dependencies in network data stream. Patterns of science data traffic are learned with a one-class neural network, which blends an Adversarial Autoencoder (AAE) with Gaussian mixture density estimation. Testing traffic flows exhibiting significant deviations from the learned baseline of normality are marked as anomalies. We trained ThunderSecure on hundreds of billions of science data packets recorded from a 100G research network at Fermi National Accelerator Laboratory. The detection performance was evaluated on traffic captured from the same research network days and weeks after the training with different types of attack flows injected. To the best of our knowledge, this is the first ML work in the area of network anomaly detection that has been validated on such extreme scale datasets. Results show that ThunderSecure can recognize science data traffic that are captured long after the training and make nearly certain detection on those with anomalous flows injected, even in the case when the anomaly-to-normal mixing ratio is 0.1%.

Gong, Qian↗

Performance and usability enhancements for continuous subgraph matching queries on graph-structured data

A query graph, which includes vertices and edges, represents a query on graph-structured data. The query graph is decomposed into query subgraphs. A network analysis tool performs continuous subgraph matching queries to facilitate analysis of computer network traffic, social media events, or other streams of data represented as a dynamic data graph (graph-structured data). This can help identify emerging trends in the data. Some features of the network analysis tool enhance performance by effectively utilizing distributed computing resources (including processing cores and memory at different nodes of a cluster) to speed up the process of updating the dynamic data graph and detecting matches of query subgraphs. Features of a query graph building tool enhance usability by providing intuitive ways to specify query graphs and their subgraphs. Features of a results visualization tool enhance usability by providing an intuitive way to present the results of continuous subgraph matching queries.

Choudhury, Sutanay↗

A Modularized Urban Scale Building Energy Modeling Framework Designed with An Open Mind

In recent years, physics-based building energy modeling (BEM) has started being used to evaluate the performance of buildings in the context of connected communities and on an urban scale to study their aggregated energy use, interactions, and impacts on the energy supply infrastructure and environment. The development of urban-scale BEM solutions needs extensive effort. Existing attempts tend to focus on different aspects of BEM on an urban scale, such as collecting as-built building data from different information sources, integrating geometry modeling with geographic information systems (GISs), representing operational and occupancy profiles, automating workflow, processing and visualizing the results, and conducting large-scale simulations. Urban-scale BEM development would benefit from multi-disciplinary research areas and from an open platform to adopt advancements on data sources and tools. For these purposes, this research proposes a modularized bottom-up model creation and simulation framework that is built on the state-of-the-art BEM tools and can accommodate different building stock data. This framework uses a standardized schema to describe building design and operational characteristics, and it can be instantiated from different building survey datasets with heterogeneous structures. The paper demonstrates how thousands of surveyed buildings from the 2012 U.S. Energy Information Administration’s Commercial Buildings Energy Consumption Survey (CBECS) were one-to-one converted to EnergyPlus models through the schema and the model generation process, then simulated with distributed computing, and their results are summarized.

Lei, Xuechen↗

Minimizing Cr-Evaporation from Balance of Plant Components by Utilizing Cost-Effective Alumina-Forming Austenitic Steels

A solid oxide fuel cell (SOFC) is a clean and efficient energy conversion device. The development of intermediate-temperature SOFCs has made it preferable to use metallic interconnects (MICs) to greatly reduce the cost and significantly increase the efficiency compared to ceramic interconnect materials. However, gaseous chromium species will evaporate from the chromium-containing layer formed on the surface of commonly used MICs and balance of plant (BoP) components. Volatile chromium species have been shown to form solid deposits which poison the cathodes of SOFCs, causing drastic cell performance degradation and thereby limiting commercialization. In order to alleviate the Cr poisoning and achieve long-term high performance of SOFC stacks, various Al2O3-forming austenitic (AFA) stainless steels applied at different temperatures are evaluated in this project. Based on our Phase I results (500 h operation), it is shown that on the AFAs, an alumina-based protective layer forms under high temperature that is invulnerable to water vapor effects and suppresses the diffusion of chromium and manganese which can prevent the generation of spinels on the alloy surface. Therefore, 310S, OC4 and OC5 at 800 °C and 625, OC11 and OC11LZ at 900 °C are selected to be tested in the Phase II long-term operation. From the results, we find that besides the lower Cr evaporation rates and better oxidation resistance of AFAs than benchmark alloys after short-term (500 h) operation, AFAs also possess the sturdy and compact alumina layer after a long-term operation (5000 h). A protective oxide layer is of great importance for the long-term high-temperature operation of structural materials. The high-temperature oxidation behavior of AFAs and commercial 310S and 625 alloys in Air + 10% H 2 O at 800 °C and 900 °C is systematically investigated. Severe breakaway oxidation and minor spallation are observed for the 310S and 625 after the long-term operation, while the AFAs show high oxidation resistance. It is found that the formation of a continuous alumina layer could greatly prevent the volatilization of chromium vapors. In addition, the corresponding evolving models are discussed. Chromium evaporation from BoP components in high-temperature environment could severely deteriorate the electrochemical performance of SOFC. Several methods were applied to evaluate the Cr evaporation rates of BoP after 500 h exposure at 800 °C to 900 °C in air with 10% H 2 O. An optimal method was designed to exclude the effect of silicon (Si) deposits from quartz tube and sodium (Na) deposits from the sodium carbonate on the oxidation process and the chemical interaction between Cr gaseous species and alumina tube which could provide further quantitative correlation of the evaporated Cr species quantities and degradation rates of SOFC. Based on the great performance of AFAs after long-term operation, AFAs are assembled with Anode-supported cells (ASC) to investigate the the Cr deposition of anode-supported cell under a constant current density of 0.5 A cm -2 at 800 °C with AFA alloys compared with commercial alloys. In addition, the anodic and cathodic processes are deconvoluted by distribution of relaxation times (DRT) method which are comprehensively discussed. It is found out that the voltage exhibited a slight decrease of 5.09 % and 1.54 % in the presence of OC11 and OC11LZA alloy, respectively. However, a considerable decrease of 22.14 % and 12.06 % was determined in the presence of 310S and 625 alloy, respectively. AFAs possessing the low-cost, low Cr evaporation rates and the high oxidation resistance will make it of great potential to replace the existing BoP components.

36 MATERIALS SCIENCE↗