Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed clustering methods”

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 199 records · Page 11

Nematic order of spin structures of clusters in the hubbard model

The possibility of interpreting the normal pseudogap state of cuprates as a result of the formation of spin and charge structures is investigated for solutions of the Hubbard model of a finite 2D cluster based on the mean field method. The iterative self-consistency procedure reduces the initial uncorrelated spin distributions to stable structures. The Fourier components of the charge and spin distributions in such structures have peaks for characteristic incommensurate quasi-momenta depending on the doping. It is shown that for any doping, the density of states of the system has a sharp minimum (pseudogap) at the Fermi level. This emergence of the gap just at the Fermi level is a property typical of not only the superconducting state, but also the normal state of spin glasses. The characteristics of the Fermi surface averaged over the implemented structures and the properties of quasiparticles in the nodal and antinodal regions of the quasi-momentum are considered.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Parallel computational fluid dynamics - Implementations and results

The present volume on parallel CFD discusses implementations on parallel machines, numerical algorithms for parallel CFD, and performance evaluation and computer science issues. Attention is given to a parallel algorithm for compressible flows through rotor-stator combinations, a massively parallel Euler solver for unstructured grids, a fast scheme to analyze 3D disk airflow on a parallel computer, and a block implicit multigrid solution of the Euler equations. Topics addressed include a 3D ADI algorithm on distributed memory multiprocessors, clustered element-by-element computations for fluid flow, hypercube FFT and the Fourier pseudospectral method, and an investigation of parallel iterative algorithms for CFD. Also discussed are fluid dynamics using interface methods on parallel processors, sorting for particle flow simulation on the connection machine, a large grain mapping method, and efforts toward a Teraflops capability for CFD.

Simon, Horst D.↗

Air Quality Monitoring Case Study Using Mobile Low-cost Sensors Mounted on Trash-Trucks: Methods Development and Lessons Learned

Air quality monitoring (AQM) is crucial for cities to develop management plans supporting population health. However, there is a dearth of measurements due to the high cost of standard reference instruments. Mobile AQM using low-cost sensors deployed on routine fleets of vehicles can enable the continuous detection of fine-scale pollutant variations in cities at a lower cost. New methods need to be developed to interpret these measurements. This paper presents three such methods. First, we propose a technique to identify aerosol hotspots. Second, we employ techniques published previously to assess the generalizable map of fine and coarse particle number concentrations, to understand qualitatively the contribution of local and regional sources across the region sampled. By using the raw number concentration of differently sized particles from the Optical Particle Counters (OPCs) instead of the noisier mass concentrations, we obtain more robust results. Third, in order to evaluate source signatures in cities, we propose another technique, in which we cluster the entire range of aerosol size-distribution measurements acquired. The properties of each cluster provide insight into the aerosol source characteristics in the sampling environment. We test these methods using a dataset we collected by mounting OPCs on two trash-trucks in Cambridge, Massachusetts.

Drive-by sensing↗

Air Quality Monitoring Case Study Using Mobile Low-cost Sensors mounted on Trash-Trucks: Methods Development and Lessons Learned

Air quality monitoring (AQM) is crucial for cities to develop management plans supporting population health. However, there is a dearth of measurements due to the high cost of standard reference instruments. Mobile AQM using low-cost sensors deployed on routine fleets of vehicles can enable the continuous detection of fine-scale pollutant variations in cities at a lower cost. New methods need to be developed to interpret these measurements. This paper presents three such methods. First, we propose a technique to identify aerosol hotspots. Second, we employ techniques published previously to assess the generalizable map of fine and coarse particle number concentrations, to understand qualitatively the contribution of local and regional sources across the region sampled. By using the raw number concentration of differently sized particles from the Optical Particle Counters (OPCs) instead of the noisier mass concentrations, we obtain more robust results. Third, in order to evaluate source signatures in cities, we propose another technique, in which we cluster the entire range of aerosol size-distribution measurements acquired. The properties of each cluster provide insight into the aerosol source characteristics in the sampling environment. We test these methods using a dataset we collected by mounting OPCs on two trash-trucks in Cambridge, Massachusetts.

Drive-by sensing↗

Development of fracture diagnostic methods for fluid distribution based on quantitative interpretation of distributed acoustic sensing and distributed temperature sensing

Multistage hydraulic fracturing design on horizontal wells has significantly evolved with larger fluid volume, more fracturing stages, and tighter perforation cluster spacing to efficiently stimulate unconventional reservoirs. From the published field observations, the recent fracturing design results in complex fracture networks or swarm of fractures. Fracture treatment evaluation is extremely challenging in such a case because of the large amount of variables in well completion and stimulation design. Combined measurements from different technologies can help in fracture diagnosis. Fluid distribution, either during fracture injection or during production, directly relates to the stimulation efficiency at the cluster level and at the stage level. Because it is unlikely in the real world to distribute the injected fluid uniformly among all the clusters, we need diagnostic techniques to generate the flow profile along a lateral. Fiber-optic measurements, such as distributed acoustic sensing (DAS) and distributed temperature sensing (DTS), are currently used to diagnose downhole flow conditions. This technology allows us to qualitatively confirm the fluid flow profile and other issues occurring downhole during fracturing such as leakage through plugs. For optimizing a fracturing design, we also need to understand how the design parameters are correlated with the stimulation efficiency. In this study, we combine two sets of models of DAS and DTS data interpretation for injected fluid volume distribution. The DAS is interpreted based on an empirical correlation between fluid flow rates and frequency band energy from the acoustic signals. The DTS is interpreted by performing temperature history match-based thermal energy conservation. Because of the completely different physics behind the interpretations, the confirmation of two interpretations provides confidence in fluid distribution.

Geochemistry & Geophysics↗

Microtomography and pore-scale modeling of two-phase Fluid Distribution

Synchrotron-based X-ray microtomography (micro CT) at the Advanced Light Source (ALS) line 8.3.2 at the Lawrence Berkeley National Laboratory produces three-dimensional micron-scale-resolution digital images of the pore space of the reservoir rock along with the spacial distribution of the fluids. Pore-scale visualization of carbon dioxide flooding experiments performed at a reservoir pressure demonstrates that the injected gas fills some pores and pore clusters, and entirely bypasses the others. Using 3D digital images of the pore space as input data, the method of maximal inscribed spheres (MIS) predicts two-phase fluid distribution in capillary equilibrium. Verification against the tomography images shows a good agreement between the computed fluid distribution in the pores and the experimental data. The model-predicted capillary pressure curves and tomography-based porosimetry distributions compared favorably with the mercury injection data. Thus, micro CT in combination with modeling based on the MIS is a viable approach to study the pore-scale mechanisms of CO{sub 2} injection into an aquifer, as well as more general multi-phase flows.

54 ENVIRONMENTAL SCIENCES↗

Protein structure prediction enhanced with evolutionary diversity : SPEED.

For naturally occurring proteins, similar sequence implies similar structure. Consequently, multiple sequence alignments (MSAs) often are used in template-based modeling of protein structure and have been incorporated into fragment-based assembly methods. Our previous homology-free structure prediction study introduced an algorithm that mimics the folding pathway by coupling the formation of secondary and tertiary structure. Moves in the Monte Carlo procedure involve only a change in a single pair of {phi},{psi} backbone dihedral angles that are obtained from a Protein Data Bank-based distribution appropriate for each amino acid, conditional on the type and conformation of the flanking residues. We improve this method by using MSAs to enrich the sampling distribution, but in a manner that does not require structural knowledge of any protein sequence (i.e., not homologous fragment insertion). In combination with other tools, including clustering and refinement, the accuracies of the predicted secondary and tertiary structures are substantially improved and a global and position-resolved measure of confidence is introduced for the accuracy of the predictions. Performance of the method in the Critical Assessment of Structure Prediction (CASP8) is discussed.

59 BASIC BIOLOGICAL SCIENCES↗

The Architecture of Metabolism Maximizes Biosynthetic Diversity in the Largest Class of Fungi

Abstract Ecological diversity in fungi is largely defined by metabolic traits, including the ability to produce secondary or “specialized” metabolites (SMs) that mediate interactions with other organisms. Fungal SM pathways are frequently encoded in biosynthetic gene clusters (BGCs), which facilitate the identification and characterization of metabolic pathways. Variation in BGC composition reflects the diversity of their SM products. Recent studies have documented surprising diversity of BGC repertoires among isolates of the same fungal species, yet little is known about how this population-level variation is inherited across macroevolutionary timescales. Here, we applied a novel linkage-based algorithm to reveal previously unexplored dimensions of diversity in BGC composition, distribution, and repertoire across 101 species of Dothideomycetes, which are considered the most phylogenetically diverse class of fungi and known to produce many SMs. We predicted both complementary and overlapping sets of clustered genes compared with existing methods and identified novel gene pairs that associate with known secondary metabolite genes. We found that variation among sets of BGCs in individual genomes is due to nonoverlapping BGC combinations and that several BGCs have biased ecological distributions, consistent with niche-specific selection. We observed that total BGC diversity scales linearly with increasing repertoire size, suggesting that secondary metabolites have little structural redundancy in individual fungi. We project that there is substantial unsampled BGC diversity across specific families of Dothideomycetes, which will provide a roadmap for future sampling efforts. Our approach and findings lend new insight into how BGC diversity is generated and maintained across an entire fungal taxonomic class.

59 BASIC BIOLOGICAL SCIENCES↗

Optimization of tomographic reconstruction workflows on geographically distributed resources

New technological advancements in synchrotron light sources enable data acquisitions at unprecedented levels. This emergent trend affects not only the size of the generated data but also the need for larger computational resources. Although beamline scientists and users have access to local computational resources, these are typically limited and can result in extended execution times. Applications that are based on iterative processing as in tomographic reconstruction methods require high-performance compute clusters for timely analysis of data. Here, time-sensitive analysis and processing of Advanced Photon Source data on geographically distributed resources are focused on. Two main challenges are considered: (i) modeling of the performance of tomographic reconstruction workflows and (ii) transparent execution of these workflows on distributed resources. For the former, three main stages are considered: (i) data transfer between storage and computational resources, (i) wait/queue time of reconstruction jobs at compute resources, and (iii) computation of reconstruction tasks. These performance models allow evaluation and estimation of the execution time of any given iterative tomographic reconstruction workflow that runs on geographically distributed resources. For the latter challenge, a workflow management system is built, which can automate the execution of workflows and minimize the user interaction with the underlying infrastructure. The system utilizes Globus to perform secure and efficient data transfer operations. The proposed models and the workflow management system are evaluated by using three high-performance computing and two storage resources, all of which are geographically distributed. Workflows were created with different computational requirements using two compute-intensive tomographic reconstruction algorithms. Experimental evaluation shows that the proposed models and system can be used for selecting the optimum resources, which in turn can provide up to 3.13× speedup (on experimented resources). Furthermore, the error rates of the models range between 2.1 and 23.3% (considering workflow execution times), where the accuracy of the model estimations increases with higher computational demands in reconstruction tasks.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

A parallel, distributed memory implementation of the adaptive sampling configuration interaction method

The many-body simulation of quantum systems is an active field of research that involves several different methods targeting various computing platforms. Many methods commonly employed, particularly coupled cluster methods, have been adapted to leverage the latest advances in modern high-performance computing. Selected configuration interaction (sCI) methods have seen extensive usage and development in recent years. However, the development of sCI methods targeting massively parallel resources has been explored only in a few research works. Here, we present a parallel, distributed memory implementation of the adaptive sampling configuration interaction approach (ASCI) for sCI. In particular, we will address the key concerns pertaining to the parallelization of the determinant search and selection, Hamiltonian formation, and the variational eigenvalue calculation for the ASCI method. Load balancing in the search step is achieved through the application of memory-efficient determinant constraints originally developed for the ASCI-PT2 method. The presented benchmarks demonstrate near optimal speedup for ASCI calculations of Cr 2 (24e, 30o) with 10 6 , 10 7 , and 3 × 10 8 variational determinants on up to 16 384 CPUs. Importantly, to the best of the authors’ knowledge, this is the largest variational ASCI calculation to date.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

A Preliminary Study of In-House Monte Carlo Simulations: An Integrated Monte Carlo Verification System

Purpose: To develop an infrastructure for the integrated Monte Carlo verification system (MCVS) to verify the accuracy of conventional dose calculations, which often fail to accurately predict dose distributions, mainly due to inhomogeneities in the patient's anatomy, for example, in lung and bone. Methods and Materials: The MCVS consists of the graphical user interface (GUI) based on a computational environment for radiotherapy research (CERR) with MATLAB language. The MCVS GUI acts as an interface between the MCVS and a commercial treatment planning system to import the treatment plan, create MC input files, and analyze MC output dose files. The MCVS consists of the EGSnrc MC codes, which include EGSnrc/BEAMnrc to simulate the treatment head and EGSnrc/DOSXYZnrc to calculate the dose distributions in the patient/phantom. In order to improve computation time without approximations, an in-house cluster system was constructed. Results: The phase-space data of a 6-MV photon beam from a Varian Clinac unit was developed and used to establish several benchmarks under homogeneous conditions. The MC results agreed with the ionization chamber measurements to within 1%. The MCVS GUI could import and display the radiotherapy treatment plan created by the MC method and various treatment planning systems, such as RTOG and DICOM-RT formats. Dose distributions could be analyzed by using dose profiles and dose volume histograms and compared on the same platform. With the cluster system, calculation time was improved in line with the increase in the number of central processing units (CPUs) at a computation efficiency of more than 98%. Conclusions: Development of the MCVS was successful for performing MC simulations and analyzing dose distributions.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

A CHEMICAL ABUNDANCE STUDY OF 10 OPEN CLUSTERS BASED ON WIYN -HYDRA SPECTROSCOPY

We present a detailed chemical abundance study of evolved stars in 10 open clusters based on Hydra multi-object echelle spectra obtained with the WIYN 3.5 m telescope. From an analysis of both equivalent widths and spectrum synthesis, abundances have been determined for the elements Fe, Na, O, Mg, Si, Ca, Ti, Ni, Zr, and for two of the 10 clusters, Al and Cr. To our knowledge, this is the first detailed abundance analysis for clusters NGC 1245, NGC 2194, NGC 2355, and NGC 2425. These 10 clusters were selected for analysis because they span a Galactocentric distance range R{sub gc} {approx} 9-13 kpc, the approximate location of the transition between the inner and outer disks. Combined with cluster samples from our previous work and those of other studies in the literature, we explore abundance trends as a function of cluster R{sub gc}, age, and [Fe/H]. As found previously by us and other studies, the [Fe/H] distribution appears to decrease with increasing R{sub gc} to a distance of {approx}12 kpc and then flattens to a roughly constant value in the outer disk. Cluster average element [X/Fe] ratios appear to be independent of R{sub gc}, although the picture for [O/Fe] is more complicated with a clear trend of [O/Fe] with [Fe/H] and sample incompleteness. Other than oxygen, no other element [X/Fe] exhibits a clear trend with [Fe/H]; likewise, there does not appear to be any strong correlation between abundance and cluster age. We divided clusters into different age bins to explore temporal variations in the radial element distributions. The radial metallicity gradient appears to have flattened slightly as a function of time, as found by other studies. There is also some indication that the transition from the inner disk metallicity gradient to the {approx}constant [Fe/H] distribution of the outer disk occurs at different Galactocentric radii for different age bins. However, interpretation of the time evolution of radial abundance distributions is complicated by the unequal R{sub gc} and [Fe/H] ranges spanned by clusters in different age bins.

79 ASTRONOMY AND ASTROPHYSICS↗

Evolution of local atomic structure during solidification of Al2Au liquid: An ab initio study

The local atomic structure evolution in Al2Au alloy during solidification from 2000 K to 400 K was studied by ab initio molecular dynamics simulations and analyzed using the structure factor, pair correlation functions, bond angle distributions, the Honeycutt-Anderson (HA) index and Voronoi tessellation methods. It was found that the icosahedral-like clusters are negligible in the Al2Au stable liquid and supercooled liquid states, and the most abundant clusters are those having HA indices of 131 and 120 or Voronoi indices of < 0,4,4,0 >, < 0,3, 6,0 > and < 0,4,4,2 > with coordination numbers of 8, 9 and 10, respectively. These clusters are similar to the local atomic structures in the CaF2-type Al2Au crystal, revealing the existence of structure heredity between liquid and crystalline phase in Al2Au alloy. (C) 2014 Acta Materialia Inc. Published by Elsevier Ltd. All rights reserved.

36 MATERIALS SCIENCE↗

STUDYING INTERCLUSTER GALAXY FILAMENTS THROUGH STACKING gmBCG GALAXY CLUSTER PAIRS

We present a method to study the photometric properties of galaxies in filaments by stacking the galaxy populations between pairs of galaxy clusters. Using Sloan Digital Sky Survey data, this method can detect the intercluster filament galaxy overdensity with a significance of {approx}5{sigma} out to z = 0.40. Using this approach, we study the g - r color and luminosity distribution of filament galaxies as a function of redshift. Consistent with expectation, filament galaxies are bimodal in their color distribution and contain a larger blue galaxy population than clusters. Filament galaxies are also generally fainter than cluster galaxies. More interestingly, the observed filament population seems to show redshift evolution at 0.12 < z < 0.40: the blue galaxy fraction has a trend to increase at higher redshift; such evolution is parallel to the ''Butcher-Oemler effect'' of galaxy clusters. We test the dependence of the observed filament density on the richness of the cluster pair: richer clusters are connected by higher density filaments. We also test the spatial dependence of filament galaxy overdensity: this quantity decreases when moving away from the intercluster axis between a cluster pair. This method provides an economical way to probe the photometric properties of filament galaxies and should prove useful for upcoming projects like the Dark Energy Survey.

79 ASTRONOMY AND ASTROPHYSICS↗

NEW TESTS FOR DISRUPTION MECHANISMS OF STAR CLUSTERS: METHODS AND APPLICATION TO THE ANTENNAE GALAXIES

We present new tests for disruption mechanisms of star clusters based on the bivariate mass-age distribution g(M, tau). In particular, we derive formulae for g(M, tau) for two idealized models in which the rate of disruption depends on the masses of the clusters and one in which it does not. We then compare these models with our Hubble Space Telescope observations of star clusters in the Antennae galaxies over the mass-age domain in which we can readily distinguish clusters from individual stars: tau approx< 10{sup 7}(M/10{sup 4} M{sub sun}){sup 1.3} yr. We find that the models with mass-dependent disruption are poor fits to the data, even with complete freedom to adjust several parameters, while the model with mass-independent disruption is a good fit. The successful model has the simple form g(M, tau) propor to M {sup -2}tau{sup -1}, with power-law mass and age distributions, dN/dM propor to M {sup -2} and dN/dtau propor to tau{sup -1}. The predicted luminosity function is also a power law, dN/dL propor to L {sup -2}, in good agreement with our observations of the Antennae clusters. The similarity of the mass functions of star clusters and molecular clouds indicates that the efficiency of star formation in the clouds is roughly independent of their masses. The age distribution of the massive young clusters is plausibly explained by the following combination of disruption mechanisms: (1) removal of interstellar material by stellar feedback, tau approx< 10{sup 7} yr; (2) continued stellar mass loss, 10{sup 7} yr approx< tau approx< 10{sup 8} yr; (3) tidal disturbances by passing molecular clouds, tau approx> 10{sup 8} yr. None of these processes is expected to have a strong dependence on mass, consistent with our observations of the Antennae clusters. We speculate that this simple picture also applies-at least approximately-to the clusters in many other galaxies.

79 ASTRONOMY AND ASTROPHYSICS↗

Modelling the flux distribution function of the extragalactic gamma-ray background from dark matter annihilation

The one-point function (i.e., the isotropic flux distribution) is a complementary method to (anisotropic) two-point correlations in searches for a gamma-ray dark matter annihilation signature. Using analytical models of structure formation and dark matter halo properties, we compute the gamma-ray flux distribution due to annihilations in extragalactic dark matter halos, as it would be observed by the Fermi Large Area Telescope. Combining the central limit theorem and Monte Carlo sampling, we show that the flux distribution takes the form of a narrow Gaussian of 'diffuse' light, with an 'unresolved point source' power-law tail as a result of bright halos. We argue that this background due to dark matter constitutes an irreducible and significant background component for point-source annihilation searches with galaxy clusters and dwarf spheroidal galaxies, modifying the predicted signal-to-noise ratio. A study of astrophysical backgrounds to this signal reveals that the shape of the total gamma-ray flux distribution is very sensitive to the contribution of a dark matter component, allowing us to forecast promising one-point upper limits on the annihilation cross-section. We show that by using the flux distribution at only one energy bin, one can probe the canonical cross-section required for explaining the relic density, for dark matter of masses around tens of GeV.

79 ASTRONOMY AND ASTROPHYSICS↗

GALAXY CLUSTERING AND PROJECTED DENSITY PROFILES AS TRACED BY SATELLITES IN PHOTOMETRIC SURVEYS: METHODOLOGY AND LUMINOSITY DEPENDENCE

We develop a new method which measures the projected density distribution w{sub p} (r{sub p} )n of photometric galaxies surrounding a set of spectroscopically identified galaxies and simultaneously the projected cross-correlation function w{sub p} (r{sub p} ) between the two populations. In this method, we are able to divide the photometric galaxies into subsamples in luminosity intervals even when redshift information is unavailable, enabling us to measure w{sub p} (r{sub p} )n and w{sub p} (r{sub p} ) as a function of not only the luminosity of the spectroscopic galaxy, but also that of the photometric galaxy. Extensive tests show that our method can measure w{sub p} (r{sub p} ) in a statistically unbiased way. The accuracy of the measurement depends on the validity of the assumption inherent to the method that the foreground/background galaxies are randomly distributed and are thus uncorrelated with those galaxies of interest. Therefore, our method can be applied to the cases where foreground/background galaxies are distributed in large volumes, which is usually valid in real observations. We have applied our method to data from the Sloan Digital Sky Survey (SDSS) including a sample of 10{sup 5} luminous red galaxies at z {approx} 0.4 and a sample of about half a million galaxies at z {approx} 0.1, both of which are cross-correlated with a deep photometric sample drawn from the SDSS. On large scales, the relative bias factor of galaxies measured from w{sub p} (r{sub p} ) at z {approx} 0.4 depends on luminosity in a manner similar to what is found for those at z {approx} 0.1, which are usually probed by autocorrelations of spectroscopic samples in previous studies. On scales smaller than a few Mpc and at both z {approx} 0.4 and z {approx} 0.1, the photometric galaxies of different luminosities exhibit similar density profiles around spectroscopic galaxies at fixed luminosity and redshift. This provides clear observational support for the assumption commonly adopted in halo occupation distribution models that satellite galaxies of different luminosities are distributed in a similar way, following the dark matter distribution within their host halos.

79 ASTRONOMY AND ASTROPHYSICS↗

Temporal Methods to Detect Content-Based Anomalies in Social Media

Here, we develop a method for time-dependent topic tracking and meme trending in social media. Our objective is to identify time periods whose content differs signifcantly from normal, and we utilize two techniques to do so. The first is an information-theoretic analysis of the distributions of terms emitted during different periods of time. In the second, we cluster documents from each time period and analyze the tightness of each clustering. We also discuss a method of combining the scores created by each technique, and we provide ample empirical analysis of our methodology on various Twitter datasets.

97 MATHEMATICS AND COMPUTING↗