Clusters Of Different Sizes Research Articles

A Cost Model for SPARK SQL

In this paper, we propose a novel cost model for Spark SQL. The cost model covers the class of Generalized Projection, Selection, Join (GPSJ) queries. The cost model keeps into account the network and IO costs as well as the most relevant CPU costs. The execution cost is computed starting from a physical plan produced by Spark. The set of operations adopted by Spark when executing a GPSJ query are analytically modeled based on the cluster and application parameters, together with a set of database statistics. Experimental results carried out on three benchmarks and on two clusters of different sizes and with different computation features show that our model can estimate the actual execution time with about the 20 percent of errors on the average. Such an accuracy is good enough to let the system choose the most effective plan even when the execution time differences are limited. The error can be reduced to 14 percent, if the analytic model is coupled with our straggler handling strategy.

Ab initio study of interstitial helium clusters in 3C-SiC

Ab initio calculations were carried out to investigate the behaviors of interstitial Hen clusters in 3C-SiC. Based on the simulation and experimental results of SiC, we considered two types of tetrahedral sites for He atoms to be allocated, forming an interstitial Hen cluster. The compact structures of Hen clusters that were alternately surrounded by the tetrahedral sites neighboring four Si atoms (TSi) and the tetrahedral sites neighboring four C atoms (TC) were investigated, and the corresponding number of tetrahedral sites in each TSi or TC layer was determined by an empirical formula. To better illustrate the interaction between He atoms, we defined one and two types of ground states when calculating the binding energies at high and low temperatures, respectively. The average binding energies of each He atom in Hen clusters of different sizes were obtained, and the maximum binding energy of a He atom in the outermost layer of a Hen cluster was estimated to be 0.61 eV via the number of nearest neighboring and next nearest neighboring He atoms. It is of interest to note that the self-trapping mechanism was not observed in 3C-SiC, since He atoms remained in their respective tetrahedral sites at low temperatures due to their high migration barrier and weak binding energy. In contrast, the compact Hen clusters tended to dissolve at high temperatures because the He atoms located at TC sites tended to migrate into TSi sites.

Numerical analysis of heat removal from gas phase clusters during condensation of Cu nanoparticles under Ar atmosphere

Abstract The statistical analysis of the molecular dynamics modeling for the nucleation of copper (Cu) vapors under argon (Ar) atmosphere in the evaporation-condensation technology was carried out. The model was used to calculate the nucleation in a cell containing 5000 atoms of Cu and 10,000 atoms of Ar. The probabilities of various outcomes in the collision of Ar and Cu atoms with Cu clusters were determined. It was found that pair collisions have a greater probability than the triple collisions. The heat removal from clusters during a collision with Ar and Cu atoms was analyzed to find the probabilities of formation of long-lived small clusters in a variety of conditions, attempting to estimate the magnitude of heat transfer from the clusters of different sizes. It is found that the pair collisions caused a very excited dimer, whereas, the triple collisions formed a more stable dimer.

Influence of Ag, Au and Pd noble metals doping on structural, optical and antimicrobial properties of zinc oxide and titanium dioxide nanomaterials

Oxide materials (ZnO, TiO2) doped with noble metals were synthesized using the combustion technique. The results of the addition of Ag, Au, and Pd up to a concentration of 2 mol% on the structural, optical, morphological and antimicrobial properties was considered. X-ray diffraction experiments revealed that the crystal structure of the host materials remained unaltered despite doping with noble metals. From the scanning electron microscopy results, it was evident that the doped nanoparticles aggregated in clusters of different sizes in the host matrix. The plasmonic effect was also observed in the absorbance spectra of the different doped materials. The obtained materials have shown promising antimicrobial features. All ZnO materials exhibited a high antimicrobial activity, with very low minimum inhibitory concentration values, against the planktonic growth of all tested Gram-positive and Gram-negative bacterial strains. All doped materials exhibited very good anti-biofilm activity, the lowest minimal biofilm eradication concentration values being registered for ZnO doped with Au and Pd toward Escherichia coli and for ZnO doped with Ag against Candida albicans. These results indicate the potential that these materials have for antimicrobial applications in the fields of biomedicine and environmental protection.

Simultaneous spatiotemporal tracking and oxygen sensing of transient implants in vivo using hot-spot MRI and machine learning

A varying oxygen environment is known to affect cellular function in disease as well as activity of various therapeutics. For transient structures, whether they are unconstrained therapeutic transplants, migrating cells during tumor metastasis, or cell populations induced by an immunological response, the role of oxygen in their fate and function is known to be pivotal albeit not well understood in vivo. To address such a challenge in the case of generation of a bioartificial pancreas, we have combined fluorine magnetic resonance imaging and unsupervised machine learning to monitor over time the spatial arrangement and the oxygen content of implants encapsulating pancreatic islets that are unconstrained in the intraperitoneal (IP) space of healthy and diabetic mice. Statistically significant trends in the postimplantation temporal dependence of oxygen content between aggregates of 0.5-mm or 1.5-mm alginate microcapsules were identified in vivo by looking at their dispersity as well as arrangement in clusters of different size and estimating oxygen content on a pixel-by-pixel basis from thousands of 2D images. Ultimately, we found that this dependence is stronger for decreased implant capsule size consistent with their tendency to also induce a larger immunological response. Beyond the bioartificial pancreas, this work provides a framework for the simultaneous spatiotemporal tracking and oxygen sensing of other cell populations and biomaterials that change over time to better understand and improve therapeutic design across diverse applications such as cellular transplant therapy, treatments preventing metastatic formation, and modulators for improving immunologic response, for all of which oxygen is a major mechanistic component.

Crime pattern recognition in Chicago city using hadoop multinode cluster

Crimes are a social annoyance and harm our society very much in many ways. In this paper we have designed a Map Reduce algorithm to aid in the process of identification of crimes in Chicago city at different levels of abstractions. Analysis is divided into three phases. Twenty-nine different kinds of crimes have been identified in Chicago city in the first phase. In the second phase crimes are segregated into four zones namely east, west, south and north. In the third phase frequency of crime at specific location is determined. We have used approximately 40 GB of crime data from Chicago Crime Portal. The finding of this research is helpful for law enforcement officers, police and other law practitioners to improve the productivity and substantially sublime the policies so that a significant decrease in crime can be seen. The paper also compares the efficiency of multimode hadoop clusters of different sizes.

Industrial Symbiosis: towards a design process for eco-industrial clusters by integrating Circular Economy and Industrial Ecology perspectives

Industrial Symbiosis (IS) is a collective approach to competitive advantage in which separate industries create a cooperative network to exchange materials, energy, water and/or by-products. By addressing issues related to resource depletion, waste management and pollution, IS plays an important role in the transition towards sustainable development. In the literature, two conceptual perspectives on IS can be identified: the Industrial Ecology (IE) and the Circular Economy (CE) perspective. Despite the recognition of these two perspectives, their relationship remains unclear and explicit attempts to develop an integrated perspective have not been made yet. Consequently, the goal of this research is to highlight and start addressing this critical gap of knowledge in order to support future research and practice geared towards the design of new IS clusters. We pose the following research question: How can the IE and CE perspectives on IS be combined in order to support the design of IS clusters? To this end, we first investigate the two perspectives more in depth and compare them in terms of nature, features and relevance for the study of IS. This is done by applying them as conceptual lenses for the analysis of the same case study, an existing IS cluster. The comparative analysis provides insights into how the two perspectives differ, ultimately demonstrating that they are complimentary and both necessary to fully describe an IS cluster. While the CE perspective is more suitable to explain how a cluster functions from a business standpoint in the operating phase, the IE perspective is more suitable to explain its development over time and its impacts on the environment, the economy and society. Building upon the outcomes of the comparative analysis, we leverage on the discipline of Strategic Design and integrate the two perspectives into a process for designing new IS clusters. We suggest two directions for future research. First, improving our comparative analysis of the two perspectives by looking at a wider sample of IS clusters of different sizes and in different contexts. Second, focusing with more specificity on the issue of how IS clusters can be designed, potentially by trying to apply the process we propose on a real case aimed at designing a new IS cluster.

Mining the Automotive Industry: A Network Analysis of Corporate Positioning and Technological Trends

The digital transformation is driving revolutionary innovations and new market entrants threaten established sectors of the economy such as the automotive industry. Following the need for monitoring shifting industries, we present a network-centered analysis of car manufacturer web pages. Solely exploiting publicly-available information, we construct large networks from web pages and hyperlinks. The network properties disclose the internal corporate positioning of the three largest automotive manufacturers, Toyota, Volkswagen and Hyundai with respect to innovative trends and their international outlook. We tag web pages concerned with topics like e-mobility & environment or autonomous driving, and investigate their relevance in the network. Toyota and Hyundai are concerned with e-mobility throughout large parts of their web page network; Volkswagen devotes more specialized sections to it, but reveals a strong focus on autonomous driving. Sentiment analysis on individual web pages uncovers a relationship between page linking and use of positive language, particularly with respect to innovative trends. Web pages of the same country domain form clusters of different size in the network that reveal strong correlations with sales market orientation. Our approach is highly transparent, reproducible and data driven, and could be used to gain complementary insights into innovative strategies of firms and competitive landscapes.

On the intriguing emission characteristics of size tunable carbon dots derived from functionalized multi-walled carbon nanotubes

Abstract Aqueous dispersions of carbon dots (C-dots) showing size tunable multi-colours in day light are obtained through surface functionalization of multi-walled carbon nanotubes (MWCNTs). The size of the C-dots varies with the functionalization temperature and smaller C-dots are found to be formed, at higher temperatures. The average size of the C-dots synthesized at temperatures of 100 °C, 80 °C and 60 °C lies in the range 6 nm–18 nm, based on transmission electron microscopy (TEM) images. The C-dots synthesized at these temperatures exhibit blue, green and golden yellow colours in day light without being excited by any external source. The structural characterizations using X-ray diffraction, Fourier transform infrared spectroscopy and Raman spectroscopy confirm the presence of a high density of functional groups on the surfaces of the C-dots, giving rise to high aqueous solubility. Compared to the Raman spectrum of the MWCNTs, the spectrum of the C-dots shows a significant increase in the intensity of the defect related D band with respect to that of the crystalline G-band, which indicates the presence of a high concentration of defects and the formation of smaller sp2 carbon clusters. The synthesized C-dots exhibit size dependent optical absorption and emission characteristics. The intriguing optical properties of these C-dots, which include the multi-colour emission at day light and the excitation wavelength and the pH dependent photoluminescence emission, can be mainly attributed to the formation of sp2 clusters of different sizes embedded in the sp3 carbon matrix.

The interaction of ethylene with free gold cluster cations: infrared photodissociation spectroscopy combined with electronic and vibrational structure calculations

The interaction of ethylene with free gold clusters of different sizes and charge states has been previously shown theoretically to involve two different adsorption modes of the C2H4 molecule, namely: the di-σ- and π-bonded ethylene adsorption configurations. Here, we present the first experimental investigation of the structure of a series of gas-phase gold-ethylene complexes, . By employing infrared multiple-photon dissociation spectroscopy in conjunction with first-principles calculations it is revealed that up to three C2H4 molecules preferably bind to gold cations in a π-bonded configuration. The binding of all ethylene molecules is found to be dominated by partial electron donation from the ethylene molecules to the gold clusters leading to an activation of the C–C bond. The cooperative action of multiple coadsorbed C2H4 on is shown to enable additional charge back-donation and an enhanced C–C bond activation. In contrast, the strong C–H bond is not weakened and the experimental spectra do not give any indication for C–H bond dissociation. The possible correlations of the C–C bond stretch vibration with the C–C bond length and the net charge transfer are discussed.

Phase transition induced by an external field in a three-dimensional isotropic Heisenberg antiferromagnet

The critical frontier of the isotropic antiferromagnetic Heisenberg model in a magnetic field along the z-axis has been studied by mean-field and effective-field renormalization group calculations. These methods, abbreviated as MFRG and EFRG, are based on the comparison of two clusters of different sizes, each of them trying to mimic a specific Bravais lattice. The frontier line in the plane of temperature versus magnetic field was obtained for the simple cubic and the body-centered cubic lattices. Spin clusters with sizes N = 1, 2, 4 were used so as to implement MFRG-12, EFRG-12 and EFRG-24 numerical equations. For the simple cubic lattice, the MFRG frontier exhibits a notorious re-entrant behavior. This problem is improved by the EFRG technique. However, both methods agree at lower fields. For the body-centered cubic lattice, the MFRG method did not work. As in the cubic lattice, all the EFRG results agree at lower fields. Nevertheless, the EFRG-12 approach gave no solution for very low temperatures. Comparisons with other methods have been discussed.

Trends and regional variations in the administrative prevalence of attention-deficit/hyperactivity disorder among children and adolescents in Germany

There is a controversy regarding temporal trends in prevalence of attention-deficit/hyperactivity disorder (ADHD). Using nationwide claims data containing data for approximately six million statutory health insured children we aimed to examine a) trends of ADHD administrative prevalence during 2009–2016; b) regional variations in prevalence, and c) factors associated with an increased chance of ADHD diagnosis. The ICD-10 code ‘F90-hyperkinetic disorder’ was used to define an ADHD case. Global and Local Moran’s I tests were used to examine the spatial autocorrelation and k-means-cluster analysis to examine the course of ADHD prevalence in administrative districts over years. Two-level logistic regression was applied to examine individual- and district-level factors associated with ADHD diagnosis. The administrative prevalence of ADHD was 4.33% (95% CI: 4.31–4.34%). We observed pronounced small-area differences on the district level with prevalences ranging between 1.6% and 9.7%. There was evidence of strong spatial autocorrelation (Global Moran’s I: 0.46, p < 0.0001). The k-means-method identified six clusters of different size; all with a stagnating trend in the prevalence over the observation period of eight years. On the district level, a lower proportion of foreign citizens, and a higher density of paediatric psychiatrists and paediatricians were positively associated with ADHD with a more pronounced effect in urban districts.

K-Nets: Clustering through nearest neighbors networks

Importance of clustering is signified by its applications in many scientific fields and the uninterrupted presence of related algorithmic methods in the literature for more than half a century. The majority of these methods have complex operation and most of the times aim on a specific application field. Here we present K-Networks (K-Nets), a deterministic algorithm based on a network structure whose nodes represent data points and can intrinsically work within a single or multi-layer architecture. The number of interconnections of each node with the rest of the network indicates the clusteringresolution degree and can be the same or different in every layer. K-Nets automatically detects symmetric structures in data that can be utilized to recover clusters of different size, shape, or partitions composed of a specific number of clusters. We provide evidence that K-Nets performance compares favorably with other well established clustering algorithms in a wide range of fields, while it can be employed in cases when data dimensionality prohibits their application.

Effect of argon cluster ion beam on fused silica surface morphology

The gas cluster ion bombardment technique is promising for low-damage surface smoothing of optical materials. In this work, the morphology evolution of fused silica surfaces during argon cluster ion beam processing was investigated. To characterize the surface roughness, power spectral density functions (PSD) were obtained from a wide spatial frequency range of measurements. The power spectral density function dependencies on the depth of the removed material and cluster ion parameters are shown. Based on the experimental results, it is demonstrated that the use of gas clusters of different sizes and specific energy allows a fused silica surface to be smoothed in the middle spatial frequency (0.01–1 μm−1) and high spatial frequency (1–100 μm−1) ranges.

Unsupervised learning about 4D features of microparticle motion.

Material clusters of different sizes are known to exist in high-temperature plasmas due to plasma-wall interactions. The facts that these clusters, ranging from sub-microns to above mm in size, can move from one location to another quickly and that there are a lot of them make high-speed imaging and tracking one of the best, effective, and sometimes only diagnostic. An unsupervised machine learning technique based on deconvolutional neural networks is developed to analyze two-camera videos of high-temperature microparticles generated from exploding wires. The neural network utilizes a locally competitive algorithm to infer representations and optimize a dictionary composed of kernels, or basis vectors, for image analysis. Our primary goal is to use this method for feature recognition and prediction of the time-dependent three-dimensional (or "4D") microparticle motion. Features equivalent to local velocity vectors have been identified as the dictionary kernels or "building blocks" of the scene. The dictionary elements from the left and right camera views are found to be strongly correlated and satisfy the projection geometrical constraints. The results show that unsupervised machine learning techniques are promising approaches to process large sets of images for high-temperature plasmas and other scientific experiments. Machine learning techniques can be useful to handle the large amount of data and therefore aid the understanding of plasma-wall interaction.

Dynamics of reactions inhibiting epitaxial growth of Si(100) surfaces via interaction with hydrogen chloride

The dynamics of reactions of HCl at growing Si(100) surfaces was investigated through computational methods, using clusters of different sizes and shapes. The analysis was conducted through Gaussian09 at the B3LYP/6-31G(d,p) level. It was found that adsorption of HCl on H-passivated dimers does not proceed as easily as on non-passivated dimers. In addition, analysis of reactions between HCl and a SiCl2 moiety adsorbed on a surface dimer indicated that HCl can lead to detachment of the SiCl2 moiety from the surface in the form of a SiCl2 or a SiHCl3 molecule, or to diffusion of the moiety to a neighboring dimer. However, when the dimer on which SiCl2 adsorbs had been H-passivated prior to adsorption, the reaction of HCl with the SiCl2 moiety becomes significantly more difficult. H-passivation of surface dimers seems thus to promote surface growth by protecting the surface against HCl adsorption and by preventing detachment or diffusion of the adsorbed SiCl2 moiety, which is pointed out as an important intermediate in the growth mechanism. These results are in consonance with evidence found for the convenience of H-passivation of surface dimers prior to silicon deposition, already published in the literature.

Co-crystallization of atomically precise metal nanoparticles driven by magic atomic and electronic shells

This paper reports co-crystallization of two atomically precise, different-size ligand-stabilized nanoclusters, a spherical (AuAg)267(SR)80 and a smaller trigonal-prismatic (AuAg)45(SR)27(PPh3)6 in 1:1 ratio, characterized fully by X-ray crystallographic analysis (SR = 2,4-SPhMe2). The larger cluster has a four concentric-shell icosahedral structure of Ag@M12@M42@M92@Ag120(SR)80 (M = Au or Ag) with the inner-core M147 icosahedron observed here for metal nanoparticles. The cluster has an open electron shell of 187 delocalized electrons, fully metallic, plasmonic behavior, and a zero HOMO-LUMO energy gap. The smaller cluster has an 18-electron shell closing, a notable HOMO-LUMO energy gap and a molecule-like optical spectrum. This is the first direct demonstration of the simultaneous presence of competing effects (closing of atom vs. electron shells) in nanocluster synthesis and growth, working together to form a co-crystal of different-sized clusters. This observation suggests a strategy that may be helpful in the design of other nanocluster systems via co-crystallization.

Electroless Deposition of Composite Materials: From Fundamental Mechanisms to Applications

Electroless nickel phosphorus (Ni−P) plating is an autocatalytic reduction process that has been used for various engineering applications. High phosphorus metallic matrix materials produced by electroless deposition are examined to exploit the good corrosion behavior of such coatings and the high mechanical properties of the composites. In order to obtain the desired properties, it is important to deeply understand the deposition mechanism and its influence on the final characteristics of the layer. The kinetic mechanism of the electroless deposition of NiP matrix from hypophosphite solutions is investigated from first principles with the intent of determining a set of elementary reactions that could effectively identify the most relevant steps of this complex process. The outcome of the computational approach is discussed in terms of experimental results obtained by electrochemical methods and surface analysis techniques [1,2]. The interaction of Ni++ and H2PO2 - with the Ni surface was studied using Ni clusters of different sizes (from 4 to 12 atoms), introducing solvation in water through the implicit polarizable continuum model. Though several theories were proposed to describe this key step in the literature, our search was guided by the original insight published several years ago [3], according to which the reduction of Ni++ requires the formation of an intermediate species containing the NiOH group. Such species would then take active part in the transfer of the OH group to H2PO2 -, thus favoring its oxidation to H2PO3-. In order to understand completely the adsorption phenomena on the electrode surface during the electroless deposition process, in situ-Raman analyses were performed. The effect of each complexing agent dissolved into electroless NiP bath on complexes adsorbed on the electrode surface and the comparison with the DFT computational analysis will be presented, highlighting the effect of additives, e.g. tungsten salts, added to the electrolyte to modify the properties of the matrix. A tool based on computational fluid dynamics CFD simulation for the co-deposition of the dispersed phase is also presented, showing the correlation between particles’ load in the bath and the amount of co-deposited particles as a function of hydrodynamic conditions. The results are discussed in light of the theoretical models proposed to describe the co-deposition processes. Following the theoretical description, composite coatings are produced and characterized in terms of particles content and distribution, morphology and mechanical properties. Erosion resistance is specifically investigated with respect to sand particles fluxes, discussing the optimal combination between the metallic matrix and the dispersed phase. Corrosion behavior is presented both in electrochemical tests and in accelerated tests in salt spray chamber. The effect of the applied thermal treatment to modify the mechanical properties is investigated and discussed as a function of the dispersed phase. [1] P.L. Cavallotti, L. Magagnin, Influence of added elements on ACD electroless NiP, ECS Transactions, 50 (53) 1-8 (2013). [2] P.L. Cavallotti, L. Magagnin, C. Cavallotti, Influence of added elements on autocatalytic chemical deposition electroless NiP, Electrochim. Acta (2013), http://dx.doi.org/10.1016/j.electacta.2013.09.083. [3] P.L. Cavallotti, G. Salvago, Electrochimica Metallorum III, 239 (1968)

A novel and fully automatic spike-sorting implementation with variable number of features.

The most widely used spike-sorting algorithms are semiautomatic in practice, requiring manual tuning of the automatic solution to achieve good performance. In this work, we propose a new fully automatic spike-sorting algorithm that can capture multiple clusters of different sizes and densities. In addition, we introduce an improved feature selection method, by using a variable number of wavelet coefficients, based on the degree of non-Gaussianity of their distributions. We evaluated the performance of the proposed algorithm with real and simulated data. With real data from single-channel recordings, in ~95% of the cases the new algorithm replicated, in an unsupervised way, the solutions obtained by expert sorters, who manually optimized the solution of a previous semiautomatic algorithm. This was done while maintaining a low number of false positives. With simulated data from single-channel and tetrode recordings, the new algorithm was able to correctly detect many more neurons compared with previous implementations and also compared with recently introduced algorithms, while significantly reducing the number of false positives. In addition, the proposed algorithm showed good performance when tested with real tetrode recordings.NEW & NOTEWORTHY We propose a new fully automatic spike-sorting algorithm, including several steps that allow the selection of multiple clusters of different sizes and densities. Moreover, it defines the dimensionality of the feature space in an unsupervised way. We evaluated the performance of the algorithm with real and simulated data, from both single-channel and tetrode recordings. The proposed algorithm was able to outperform manual sorting from experts and other recent unsupervised algorithms.

Role of the Cluster Structure of Amorphous Fe67Cr18B15 Alloy in Magnetism and in the Changing of Electron Scattering Mechanisms under the Influence of Ion (Ar+) Irradiation

The contribution of clusters of different sizes to magnetism and the switching of electron scattering mechanisms in amorphous Fe67Cr18B15 alloy during ion Ar+ irradiation is studied. The cluster magnetism is found to be related to the presence of clusters of the following two types: large α-(Fe, Cr) clusters of size D = 150–250 A and small (D = 40–80 A) clusters in a random intercluster medium. The generation of small ferromagnetic and antiferromagnetic clusters during ion irradiation leads to the formation of cluster glass, which affects the electrical properties of the alloy and causes a magnetic frustration. The temperature dependence of the barrier height is shown to characterize the magnetic state of the alloy in low fields. On the whole, the temperature dependence of the order parameter is a universal characteristic of the system. The temperature dependence of resistivity of initial alloys in the temperature range 98–300 K (ρ(T) ∝ T2) is determined by electron scattering by quantum defects, and the transition into a ferromagnetic state is revealed when the derivative ∂ρ/∂T ∝ T is analyzed. The increase in resistivity and the relation ρ ∝ T1/2 in strongly inhomogeneous samples after irradiation at a dose Φ = 1.5 × 1018 ions/cm2 are caused by weak localization effects, and the transition to a ferromagnetic state becomes obvious when the derivative ∂ρ/∂T ∝ T–1/2 is considered. Irradiation by fluence Φ = 3 × 1018 ions/cm2 induces a giant (twofold) increase in the alloy density, restores the ferromagnetism of large clusters, decreases the resistivity by 37%, and restores the relation ρ(T) ∝ T2, which results from the overlapping of the irradiation-induced small clusters when their concentration increases and from an increase in the alloy density. The overlapping of clusters lowers the barrier height and decreases the sensitivity of the alloy to an applied field. The relation ρ(T) ∝ T2 is valid for the entire temperature range T = 2–300 K because of the partial screening of the magnetic moments of large clusters by a medium having the properties of cluster glass.