Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Synchronization”

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 937 records · Page 52

Modeling the asynchronous Jacobi method without communication delays

Asynchronous iterative methods for solving linear systems are gaining renewed interest due to the high cost of synchronization points in massively parallel codes. Historically, theory on asynchronous iterative methods has focused on asymptotic behavior, while the transient behavior remains poorly understood. In this paper, we study a model of the asynchronous Jacobi method without communication delays, which we call simplified asynchronous Jacobi. Simplified asynchronous Jacobi can be used to model asynchronous Jacobi implemented in shared memory or distributed memory with fast communication networks. Our analysis uses the idea of a propagation matrix, which is similar in concept to an iteration matrix. We show that simplified asynchronous Jacobi can continue to reduce the residual when some processes are slower than other processes. We also show that simplified asynchronous Jacobi can converge when synchronous Jacobi does not. We verify our analysis of simplified asynchronous Jacobi using results from asynchronous Jacobi implemented in shared and distributed memory.

97 MATHEMATICS AND COMPUTING↗

Improving scalability of parallel CNN training by adaptively adjusting parameter update frequency

Synchronous SGD with data parallelism, the most popular parallelization strategy for CNN training, suffers from the expensive communication cost of averaging gradients among all workers. The iterative parameter updates of SGD cause frequent communications and it becomes the performance bottleneck. In this paper, we propose a lazy parameter update algorithm that adaptively adjusts the parameter update frequency to address the expensive communication cost issue. Our algorithm accumulates the gradients if the difference of the accumulated gradients and the latest gradients is sufficiently small. Here, the less frequent parameter updates reduce the per-iteration communication cost while maintaining the model accuracy. Our experimental results demonstrate that the lazy update method remarkably improves the scalability while maintaining the model accuracy. For ResNet50 training on ImageNet, the proposed algorithm achieves a significantly higher speedup (739.6 on 2048 Cori KNL nodes) as compared to the vanilla synchronous SGD (276.6) while the model accuracy is almost not affected (<0.2% difference).

97 MATHEMATICS AND COMPUTING↗

Synchronistic preparation of fibre-like SiC and cubic-ZrO{sub 2}/SiC composite from zircon via carbothermal reduction process

Graphical abstract: Display Omitted Highlights: ► Zircon carbothermal reduction was carried out in a tailor-made device at high-temperature air atmosphere. ► Fibre-like SiC and cubic-ZrO{sub 2}/SiC composite were obtained synchronically. ► Zirconium and silicon in zircon ore was initial separated. ► [SiO{sub 4}] was mutated to fibre-like SiC, while [ZrO{sub 8}] was transformed to cubic ZrO{sub 2}. ► The SiC were surprisingly enriched in the reducing atmosphere charred coal particles layers by gas–solid reaction. -- Abstract: Fibre-like SiC and cubic-ZrO{sub 2}/SiC composite were prepared respectively from zircon with yttrium oxide addition via carbothermal reduction process at 1600 °C for 4 h in an air atmosphere furnace, where the green samples were immerged in charred coal particles inside a high-temperature enclosed corundum crucible. The reaction products were characterized by XRD, XRF, XPS and SEM. The results indicate that ZrO{sub 2} in the products was mainly existed in the form of cubic phase. The reacted samples mainly contain cubic ZrO{sub 2}, β-SiC and trace amounts of zircon, with the SiC accounting for 14.8 wt%. Furthermore, a large quantity of fibre-like SiC was surprisingly found to concentrate in the charred coal particles layers around the samples. This study obtains fibre-like SiC and cubic-ZrO{sub 2}/SiC composite synchronically from zircon via carbothermal reduction process, which also bring a value-added high-performance application for natural zircon.

36 MATERIALS SCIENCE↗

Synthesis and photocatalytic property of BiOBr/palygorskite composites

Graphical abstract: - Highlights: • BiOBr/palygorskite composite photocatalysts were prepared for the first time. • The synchronous role of adsorbability and photocatalytic oxidation was studied. • BiOBr/palygorskite composite exhibit enhanced photocatalytic activities. • The optimal ratio of [Bi]/palygorskite was identified. - Abstract: BiOBr/palygorskite composite photocatalysts were prepared by a hydrothermal method for the first time, which were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), Fourier transformed infrared spectroscopy (FTIR), N{sub 2}-sorption and UV–vis diffuse reflectance spectroscopy. The photocatalytic activity of the as-prepared samples was evaluated by decomposition of methyl orange (MO) under visible light irradiation. The results revealed that BiOBr/palygorskite composites had strong photoabsorption in the visible light region. All BiOBr/palygorskite composites possessed higher photocatalytic activity than pure BiOBr. The highest activity was obtained by S5.0 sample. The enhanced photocatalytic performance was attributed to strong photoabsorption, the synchronous role of adsorbability and photocatalytic oxidation on BiOBr/palygorskite composites and the efficient separation of photogenerated electron–hole pairs.

36 MATERIALS SCIENCE↗

A facile in-situ etching route to hollow C@SnO2 nanocomposites and their gas-sensing properties

Highlights: • Hollow C@SnO{sub 2} nanostructure has been designed via a facile in-situ etching technique.. • The C@SnO{sub 2} products possess a high surface are and a hollow carbon core. • The growth of SnO{sub 2} nanosheets and the removal of silica template can be synchronously performed. • The C@SnO{sub 2} products manifest highly sensitive and selective to n-butanol as gas-sensing sensor. - Abstract: A novel hollow carbon@SnO{sub 2} nanocomposite has been developed via a high cost-effective and facile in-situ etching technique. The as-prepared C@SnO{sub 2} product has a high surface area of about 115 m{sup 2} g{sup −1} and a hollow carbon core, where a lot of SnO{sub 2} nanosheets grow onto the carbon shell. Particularly, the silica core as the sacrificial template can be fully removed by hydrofluoric acid (HF) byproduct during the reaction process. Moreover, the deposition of SnO{sub 2} sheets and the removal of the silica template can be synchronously performed by a simple one-step process. When applied as the gas-sensing material, the C@SnO{sub 2} product manifests highly sensitive and selective to n-butanol. The large surface area, hollow nanostructure and the perfect combination of SnO{sub 2} nanosheets and carbon layer contribute to the enhanced gas-sensing performances. The method described in our work can also be employed to design other hollow metal oxides@carbon hybrids for energy storage and catalysis.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Analysis of colorimetry using the CIE-L*a*b* system and the photocatalytic activity of photochromic films

Highlights: • The films containing fibers were efficient in the degradation of methyl orange dye. • Doping of TiO{sub 2} with W allowed the band gap reduction of TiO{sub 2}. • The photochromism observed in the films was due to the synchronicity between oxides. • The films exhibited variations between colors, and different reflectance levels. - Abstract: This work aims to explain how oxide films are correlated when exposed to UV-A radiation, due to the occurrence of similar phenomena. Thus, thin films of TiO{sub 2} and TiO{sub 2} doped with H{sub 2}WO{sub 4} or Na{sub 2}WO{sub 4}·2H{sub 2}O were made by spin-coating, heat treated between 650 and 800 °C and evaluated for their optical and photocatalytic properties. The X-ray diffraction technique (XRD) was used to determine the crystalline structure of the films. The photochromism was observed through a spectrophotometer using the CIE-Lab colorimetric method. The analysis of the films photocatalytic behavior was obtained following the photodegradation in 125 ml of methyl orange dye (20 ppm). The results showed that the films presented good photochromic and photocatalytic properties due to the synchronization between the chemical and physical properties of TiO{sub 2} and tungsten.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Measurement and analysis of fast transient instabilities

In this study, a novel beam diagnostics technique to detect fast beam transient instabilities and glitches has been developed at National Synchrotron Light Source II (NSLS-II). Vertical spikes of beam size and centroid have been observed occasionally and randomly during high current operations. These spikes are believed to be caused by ion instabilities associated with vacuum activity localized in the ring. With the functionality of gated beam position monitor (BPM), the newly developed technique utilizes turn-by-turn (TBT) data to detect beam centroid glitches. Whenever one turn trajectory deviates from a predefined window, a global control event is generated to trigger an online TBT data acquisition. The data acquisition is synchronized for all BPMs around the ring. In the meantime, the bunch-by-bunch (BxB) data, which is acquired from a single dedicated feedback BPM, has also been collected for analyzing possible coupled bunch instabilities (CBI). In addition to CBI caused mainly by ion bursts, other types of fast transient motions, such as, the top-off residual disturbance and fast orbit disturbance, have been captured with the new tool. Lastly, the sources of the glitches can be approximately localized with the synchronized TBT data from multiple BPMs around the ring.

43 PARTICLE ACCELERATORS↗

Hierarchical Krylov and nested Krylov methods for extreme-scale computing

The solution of large, sparse linear systems is typically a dominant phase of computation for simulations based on partial differential equations, which are ubiquitous in scientific and engineering applications. While preconditioned Krylov methods are widely used and provide many advantages for solving sparse linear systems that do not have highly convergent, geometric multigrid solvers or specialized fast solvers, Krylov methods encounter well-known scaling difficulties for over 10,000 processor cores because each iteration requires at least one vector inner product, which in turn requires a global synchronization that scales poorly because of internode latency. To aid in overcoming these difficulties, we have developed hierarchical Krylov methods and nested Krylov methods in the PETSc library that reduce the number of global inner products required across the entire system (where they are expensive), though freely allow vector inner products across smaller subsets of the entire system (where they are inexpensive) or use inner iterations that do not invoke vector inner products at all. Nested Krylov methods are a generalization of inner-outer iterative methods with two or more layers. Hierarchical Krylov methods are a generalization of block Jacobi and overlapping additive Schwarz methods, where each block itself is solved by Krylov methods on smaller blocks. Conceptually, the hierarchy can continue recursively to an arbitrary number of levels of smaller and smaller blocks. As a specific case, we introduce the hierarchical FGMRES method, or h-FGMRES, and we demonstrate the impact of two-level h-FGMRES with a variable preconditioner on the PFLOTRAN subsurface flow application. We also demonstrate the impact of nested FGMRES, BiCGStab and Chebyshev methods. These hierarchical Krylov methods and nested Krylov methods significantly reduced overall PFLOTRAN simulation time on the Cray XK6 when using 10,000 through 224,000 cores through the combined effects of reduced global synchronization due to fewer global inner products and stronger inner hierarchical or nested preconditioners.

97 MATHEMATICS AND COMPUTING↗

Performance of asynchronous optimized Schwarz with one-sided communication

In asynchronous iterative methods on distributed-memory computers, processes update their local solutions using data from other processes without an implicit or explicit global synchronization that corresponds to advancing the global iteration counter. Here in this work, we test the asynchronous optimized Schwarz domain-decomposition iterative method using various one-sided (remote direct memory access) communication schemes with passive target completion. The results show that when one-sided communication is well-supported, the asynchronous version of optimized Schwarz can outperform the synchronous version even for perfectly balanced partitionings of the problem on a supercomputer with uniform nodes.

97 MATHEMATICS AND COMPUTING↗

Distributed Automatic Differentiation for Ptychography

Synchrotron radiation light source facilities are leading the way to ultrahigh resolution X-ray imaging. High resolution imaging is essential to understanding the fundamental structure and interaction of materials at the smallest length scale possible. Diffraction based methods achieve nanoscale imaging by replacing traditional objective lenses by pixelated area detectors and computational image reconstruction. Among these methods, ptychography is quickly becoming the standard for sub-30 nanometer imaging of extended samples, but at the expense of increasingly high data rates and volumes. This paper presents a new distributed algorithm for solving the ptychographic image reconstruction problem based on automatic differentiation. Input datasets are subdivided between multiple graphics processing units (GPUs); each subset of the problem is then solved either entirely independent of other subsets (asynchronously) or through sharing gradient information with other GPUs (synchronously). The algorithm was evaluated on simulated and real data acquired at the Advanced Photon Source, scaling up to 192 GPUs. The synchronous variant of our method outperformed an existing multi-GPU implementation in terms of accuracy while running at a comparable execution time.

distributed algorithms↗

In-Situ Blade Strain Measurements and Fatigue Analysis of a Cross-Flow Turbine Operating in a Tidal Flow

Cross-flow turbines (CFTs) are inherently unsteady devices with regards to operating principle and loading. By improving our understanding of the dynamic loading on these turbines, we hope to better inform CFT design, improve survivability, and reduce overall costs. The University of New Hampshire (UNH) and the National Renewable Energy Laboratory (NREL) collaborated on a project to instrument and test a four-bladed New Energy Corp. vertical axis cross-flow turbine in a real tidal flow. One blade from the 3.2 m diameter x 1.7 m height turbine was instrumented with eight full-bridge strain gauges along the span of the blade. The turbine was then deployed at the UNH-Atlantic Marine Energy Center (AMEC) Tidal Energy Test Site in Portsmouth, NH. Time-synchronized measurements of blade strain, inflow, thrust, rotational speed, and electrical output were obtained to characterize blade loading under various conditions. The blade strain was examined to assess the dynamic loading and conduct a fatigue analysis on the device.

blade strain↗

Hyaluronan suppresses prostate tumor cell proliferation through diminished expression of N-cadherin and aberrant growth factor receptor signaling

Hyaluronan (HA) production has been functionally implicated in prostate tumorigenesis and metastasis. We previously used prostate tumor cells overexpressing the HA synthesizing enzyme HAS3 or the clinically relevant hyaluronidase Hyal1 to show that excess HA production suppresses tumor growth, while HA turnover accelerates spontaneous metastasis from the prostate. Here, we examined pathways responsible for effects of HAS3 and Hyal1 on tumor cell phenotype. Detailed characterization of cell cycle progression revealed that expression of Hyal1 accelerated cell cycle re-entry following synchronization, whereas HAS3 alone delayed entry. Hyal1 expressing cells exhibited a significant reduction in their ability to sustain ERK phosphorylation upon stimulation by growth factors, and in their expression of the cyclin-dependent kinase inhibitor p21. In contrast, HAS3 expressing cells showed prolonged ERK phosphorylation and increased expression of both p21 and p27, in asynchronous and synchronized cultures. Changes in cell cycle regulatory proteins were accompanied by HA-induced suppression of N-cadherin, while E-cadherin expression and {beta}-catenin expression and distribution remained unchanged. Our results are consistent with a model in which excess HA synthesis suppresses cell proliferation by promoting homotypic E-cadherin mediated cell-cell adhesion, consequently signaling to elevate cell cycle inhibitor expression and suppress G1- to S-phase transition.

60 APPLIED LIFE SCIENCES↗

Triptolide inhibits TGF-β1-induced cell proliferation in rat airway smooth muscle cells by suppressing Smad signaling

Background: We have reported that triptolide can inhibit airway remodeling in a murine model of asthma via TGF-β1/Smad signaling. In the present study, we aimed to investigate the effect of triptolide on airway smooth muscle cells (ASMCs) proliferation and the possible mechanism. Methods: Rat airway smooth muscle cells were cultured and made synchronized, then pretreated with different concentration of triptolide before stimulated by TGF-β1. Cell proliferation was evaluated by MTT assay. Flow cytometry was used to study the influence of triptolide on cell cycle and apoptosis. Signal proteins (Smad2, Smad3 and Smad7) were detected by western blotting analysis. Results: Triptolide significantly inhibited TGF-β1-induced ASMC proliferation (P<0.05). The cell cycle was blocked at G1/S-interphase by triptolide dose dependently. No pro-apoptotic effects were detected under the concentration of triptolide we used. Western blotting analysis showed TGF-β1 induced Smad2 and Smad3 phosphorylation was inhibited by triptolide pretreatment, and the level of Smad7 was increased by triptolide pretreatment. Conclusions: Triptolide may function as an inhibitor of asthma airway remodeling by suppressing ASMCs proliferation via negative regulation of Smad signaling pathway. - Highlights: • In this study, rat airway smooth muscle cells were cultured and made synchronized. • Triptolide inhibited TGF-β1-induced airway smooth muscle cells proliferation. • Triptolide inhibited ASMCs proliferation via negative regulation of Smad signaling pathway.

60 APPLIED LIFE SCIENCES↗

Energy Transduction in Nitrogenase

Nitrogenase is a complicated two-component enzyme system that uses ATP binding and hydrolysis energy to achieve one of the most difficult chemical reactions in nature, the reduction of N2 to NH3. One component of the Mo-based nitrogenase system, Fe protein, delivers electrons one at a time to the second component, the catalytic MoFe protein. This process occurs through a series of synchronized events collectively called the “Fe protein cycle”. Elucidating details of the events associated with this cycle has constituted an important challenge in understanding the nitrogenase mechanism. Electron delivery is a multistep process involving three metal clusters with intra- and interprotein events. It is proposed that the first electron transfer event is a gated intraprotein transfer of one electron from the MoFe protein P-cluster to the FeMo cofactor. Measurement of the effect of osmotic pressure on the rate of this electron transfer process revealed that it is gated by protein conformational changes. This first electron transfer is activated by binding of the Fe protein containing two bound ATP molecules. The mechanism of how this protein-protein association triggers electron transfer remains unknown. The second electron transfer event is proposed to be a rapid interprotein “backfill” with transfer of one electron from the reduced Fe protein 4Fe-4S cluster to the oxidized P-cluster. In this way, electron delivery can be viewed as a case of “deficit spending”. Such a deficit-spending electron transfer process can be envisioned as a way to achieve one-direction electron flow, limiting the potential for back electron flow. Hydrolysis of two ATP molecules associated with the Fe protein occurs after the electron transfer and therefore is not used to directly drive the electron transfer. Rather, ATP hydrolysis is proposed to contribute to relaxation of the “activated” conformational state associated with the ATP form of the complex, with the free energy from ATP hydrolysis being used to pay back energy associated with component protein association and electron transfer. Release of inorganic phosphate (Pi) and protein-protein dissociation follow electron transfer and ATP hydrolysis. The ratelimiting step for the Fe protein cycle is not dissociation of the two proteins, as previously believed, but rather is release of Pi after ATP hydrolysis, which is then followed by rapid protein-protein complex dissociation. Nitrogenase is composed of two catalytic halves that do not function independently but rather exhibit anticooperative nuclear motion in which electron transfer in one-half of the complex partially inhibits electron transfer and ATP hydrolysis in the other half. Calculations indicated the existence of anticooperative interactions across the entire nitrogenase complex, suggesting a mechanism for the control of events on opposite ends of this large complex. The mechanistic necessity for this anticooperative process remains unknown. This Account presents a working model for how all of these processes work together in the nitrogenase “machine” to transduce the energy from ATP binding and hydrolysis to drive N2 reduction.

Seefeldt, Lance C.↗

Heterogeneous Composition of Oxygen-Evolving Complexes in Crystal Structures of Dark-Adapted Photosystem II

Photosystem II (PSII) is a homodimeric protein complex that catalyzes water oxidation at the oxygen-evolving complex (OEC), a heterocubanoid calcium-tetra-manganese cluster. Here, we analyze the omit electron-density peaks of the OEC’s metal ions in five X-ray free-electron laser PSII structures at a resolution between 2.15 Å and 1.95 Å. The omit peaks can be described by the total number of electrons and approximated by the variance of electron density distribution when the distributions are spherically symmetric. We show that the number of electrons of metal centers are different in the two OECs of PSII dimers, implying that either the oxidation states and/or occupancies of individual metal ions are different in the two monomers. In either case, we find that the two OECs of dark-adapted PSII dimers in crystals are not fully synchronized in the same expected S 1 state. Differences in redox states of the OEC in PSII only partially account for the observation that the electron densities integrate to a smaller number of electrons than expected. Differences between the determined and expected relative electron numbers are much larger than the estimated errors, indicating heterogeneity in the OEC composition. Here, the reported findings suggest that two OECs of dark-adapted PSII crystals are typically not fully synchronized in the S 1 state and likely generate heterogeneity in the crystallographical data of PSII intermediate structures.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Endoribonuclease-Based Two-Component Repressor Systems for Tight Gene Expression Control in Plants

Tight control and multifactorial regulation of gene expression are important challenges in genetic engineering and are critical for the development of regulatory circuits. Meeting these challenges will facilitate transgene expression regulation and support the fine-tuning of metabolic pathways to avoid the accumulation of undesired intermediates. By employing the endoribonuclease Csy4 and its recognition sequence from Pseudomonas aeruginosa and manipulating 5'UTR of mRNA, we developed a two-component expression-repression system to tightly control synthesis of transgene products. We demonstrated that this regulatory device was functional in monocotyledonous and dicotyledonous plant species, and showed that it can be used to repress transgene expression by > 400-fold and to synchronize transgene repression. In addition to tissue-specific transgene repression, this system offers stimuli-dependent expression control. Using a bioinformatics approach, we identified 54 orthologous systems from various bacteria, and then validated in planta the activity for a few of those systems, demonstrating the potential diversity of such a two-component repressor system.

CRISPR/Cas9↗

Cerium Oxyhydroxide Clusters: Formation, Structure and Reactivity

Cerium oxyhydroxide cluster anions were produced by irradiating ceric oxide particles using 355 nm laser pulses that were synchronized with pulses of nitrogen gas admitted to the irradiation chamber. The gas pulse stabilized the nascent clusters that are largely anhydrous [CexOy] ions and neutrals. These initially-formed species react with water, principally forming closed-shell (c-s) oxohydroxy species that are described by the general formula [CexOy(OH)z]-. In general, the extent of hydroxylation varies from a value of 3 OH per Ce atom when x = 1 to a value slightly greater than 1 for x > 8. The Ce3 and Ce6 species deviate significantly from this trend: the x = 3 cluster accommodates more hydroxyl moieties compared to neighboring congeners at x = 2 and x = 4. Conversely, the x = 6 cluster is significantly less hydroxylated. Density functional theory (DFT) modeling of the cluster structures show that the hydrated clusters are hydrolyzed, and contain one-to-multiple hydroxide moieties, but not datively bound water. DFT also predicts an energetic preference for formation of highly symmetric structures as the size of the clusters increases. The calculated structures indicate that the ability of the Ce3 oxyhydroxide to accommodate more extensive hydroxylation is due to a more open, hexagonal structure in which the Ce atoms can participate in multiple hydrolysis reactions. Conversely the Ce6 oxyhydroxide has an octahedral structure that is not conducive to hydrolysis. In addition to the c-s clusters, open-shell (o-s) oxyhydroxides and superoxides are also formed, and they become more prominent as the size of the clusters increases, suggesting that the larger ceria clusters have an increased ability to stabilize a non-bonding electron. The overall intensity of the clusters tends to monotonically decrease as the cluster size increases, however this trend is interrupted at Ce13, which is significantly more stable compared to neighboring congeners, suggesting formation of a dehydrated Keggin-type structure.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Large Eddies Regulate Turbulent Flux Gradients in Coupled Stable Boundary Layers

Due to strong mean wind shear, the stable boundary layer (SBL) becomes vertically coupled. In a coupled SBL, large turbulent eddies enhance cross-layer mixing and vertically mix turbulent kinetic energy. However, large gradients in momentum and heat fluxes are frequently observed, degrading the performance of Monin-Obukhov similarity theory. It is shown that increased vertical gradients of mean variables (i.e., wind speed and potential temperature) can cause flux gradients. This process is operated upon by downward penetrating large eddies with altered phase difference, contributing unevenly to fluxes and thus causing flux gradients. In the coupled SBL with small gradients in mean variables, large eddies are vertically synchronized, contributing evenly to fluxes across layers and thus causing small flux gradients. As turbulent production becomes weak and the cross-layer difference in turbulent flux transport increases, large eddies become less vertically synchronized, contributing unevenly to fluxes across layers and thus causing large flux gradients.

58 GEOSCIENCES↗