Search NASA⌕ Search

SEARCH · Search NASA

Results for “cluster computing”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
NASA NTRS2003

Experiences using OpenMP based on Computer Directed Software DSM on a PC Cluster

In this work we report on our experiences running OpenMP programs on a commodity cluster of PCs running a software distributed shared memory (DSM) system. We describe our test environment and report on the performance of a subset of the NAS Parallel Benchmarks that have been automaticaly parallelized for OpenMP. We compare the performance of the OpenMP implementations with that of their message passing counterparts and discuss performance differences.

Hess, Matthias↗
NASA NTRS2011

Collaborative Clustering for Sensor Networks

Traditionally, nodes in a sensor network simply collect data and then pass it on to a centralized node that archives, distributes, and possibly analyzes the data. However, analysis at the individual nodes could enable faster detection of anomalies or other interesting events, as well as faster responses such as sending out alerts or increasing the data collection rate. There is an additional opportunity for increased performance if individual nodes can communicate directly with their neighbors. Previously, a method was developed by which machine learning classification algorithms could collaborate to achieve high performance autonomously (without requiring human intervention). This method worked for supervised learning algorithms, in which labeled data is used to train models. The learners collaborated by exchanging labels describing the data. The new advance enables clustering algorithms, which do not use labeled data, to also collaborate. This is achieved by defining a new language for collaboration that uses pair-wise constraints to encode useful information for other learners. These constraints specify that two items must, or cannot, be placed into the same cluster. Previous work has shown that clustering with these constraints (in isolation) already improves performance. In the problem formulation, each learner resides at a different node in the sensor network and makes observations (collects data) independently of the other learners. Each learner clusters its data and then selects a pair of items about which it is uncertain and uses them to query its neighbors. The resulting feedback (a must and cannot constraint from each neighbor) is combined by the learner into a consensus constraint, and it then reclusters its data while incorporating the new constraint. A strategy was also proposed for cleaning the resulting constraint sets, which may contain conflicting constraints; this improves performance significantly. This approach has been applied to collaborative clustering of seismic and infrasonic data collected by the Mount Erebus Volcano Observatory in Antarctica. Previous approaches to distributed clustering cannot readily be applied in a sensor network setting, because they assume that each node has the same view of the data set. A view is the set of features used to represent each object. When a single data set is partitioned across several computational nodes, distributed clustering works; all objects have the same view. But when the data is collected from different locations, using different sensors, a more flexible approach is needed. This approach instead operates in situations where the data collected at each node has a different view (e.g., seismic vs. infrasonic sensors), but they observe the same events. This enables them to exchange information about the likely cluster membership relations between objects, even if they do not use the same features to represent the objects.

Wagstaff. Loro :/↗
NASA NTRS2000

Dynamic Load-Balancing for Distributed Heterogeneous Computing of Parallel CFD Problems

The developed methodology is aimed at improving the efficiency of executing block-structured algorithms on parallel, distributed, heterogeneous computers. The basic approach of these algorithms is to divide the flow domain into many sub- domains called blocks, and solve the governing equations over these blocks. Dynamic load balancing problem is defined as the efficient distribution of the blocks among the available processors over a period of several hours of computations. In environments with computers of different architecture, operating systems, CPU speed, memory size, load, and network speed, balancing the loads and managing the communication between processors becomes crucial. Load balancing software tools for mutually dependent parallel processes have been created to efficiently utilize an advanced computation environment and algorithms. These tools are dynamic in nature because of the chances in the computer environment during execution time. More recently, these tools were extended to a second operating system: NT. In this paper, the problems associated with this application will be discussed. Also, the developed algorithms were combined with the load sharing capability of LSF to efficiently utilize workstation clusters for parallel computing. Finally, results will be presented on running a NASA based code ADPAC to demonstrate the developed tools for dynamic load balancing.

Ecer, A.↗
NASA NTRS1986

The extinction law in the open cluster NGC 457 and the intrinsic energy distribution of Phi Cassiopeiae (F0 Ia)

Five early B-type stars near the main-sequence turnoff in NGC 457 have been observed at low dispersion with the short-wavelength prime and the long-wavelength redundant cameras of the IUE satellite. The equivalent widths of spectral features that are particularly strong and sensitive to temperature and luminosity were computed in the cluster stars and in 20 lightly reddened stars of types O9-B3 and luminosity classes III-V. The comparison of the equivalent widths provides a reliable method for finding matching pairs. Having identified the best comparison star for each program star, binned fluxes were used to determine the mean extinction curve. In order to cover the visible region, monochromatic fluxes of Phi Cas were derived from observations with the intensified Reticon scanner mounted on the No. 2 0.9 m telescope of KPNO, and they were dereddened with the mean extinction curve of Savage and Mathis. Thus, the intrinsic energy distribution of Phi Cas were determined from 1500 to 5800 A for use in a detailed model-atmosphere analysis.

Rosenzweig, P.↗
NASA NTRS2024

DNA Break Clustering as a Predictor of Cell Death across Various Radiation Qualities: Influence of Cell Size, Cell Asymmetry, and Beam Orientation

Cosmic radiation, composed of high charge and energy (HZE) particles, causes cellular DNA damage that can result in cell death or mutation that can evolve into cancer. In this work, a cell death model is applied to several cell lines exposed to HZE ions spanning a broad range of linear energy transfer (LET) values. We hypothesize that chromatin movement leads to the clustering of multiple double strand breaks (DSB) within one radiation-induced foci (RIF). The survival probability of a cell population is determined by averaging the survival probabilities of individual cells, which is function of the number of pairwise DSB interactions within RIF. The simulation code RITCARD was used to compute DSB. Two clustering approaches were applied to determine the number of RIF per cell. RITCARD outputs were combined with experimental data from four normal human cell lines to derive the model parameters and expand its predictions in response to ions with LET ranging from ∼0.2keV/μmto∼3000keV/μm. Spherical and ellipsoidal nuclear shapes and two ion beam orientations were modeled to assess the impact of geometrical properties on cell death. The calculated average number of RIF per cell reproduces the saturation trend for high doses and high-LET values that is usually experimentally observed. The cell survival model generates the recognizable bell shape of LET dependence for the relative biological effectiveness (RBE). At low LET, smaller nuclei have lower survival due to increased DNA density and DSB clustering. At high LET, nuclei with a smaller irradiation area either because of a smaller size or a change in beam orientation have a higher survival rate due to a change in the distribution of DSB/RIF per cell. If confirmed experimentally, the geometric characteristics of cells would become a significant factor in predicting radiation-induced biological effects.

cell survival↗
NASA NTRS2002

A Dissimilarity Measure for Clustering High- and Infinite Dimensional Data that Satisfies the Triangle Inequality

The cosine or correlation measures of similarity used to cluster high dimensional data are interpreted as projections, and the orthogonal components are used to define a complementary dissimilarity measure to form a similarity-dissimilarity measure pair. Using a geometrical approach, a number of properties of this pair is established. This approach is also extended to general inner-product spaces of any dimension. These properties include the triangle inequality for the defined dissimilarity measure, error estimates for the triangle inequality and bounds on both measures that can be obtained with a few floating-point operations from previously computed values of the measures. The bounds and error estimates for the similarity and dissimilarity measures can be used to reduce the computational complexity of clustering algorithms and enhance their scalability, and the triangle inequality allows the design of clustering algorithms for high dimensional distributed data.

Socolovsky, Eduardo A.↗
NASA NTRS1978

Predicting the workload of a computer system

A technique based on clustering is described for characterizing the current workload of a computer system and for predicting future workload. The technique was applied to predicting the load for an installation operating two machines processing about 20,000 job steps per month.

Agrawala, A. K.↗
NASA NTRS1981

An algorithm for spatial heirarchy clustering

A method for utilizing both spectral and spatial redundancy in compacting and preclassifying images is presented. In multispectral satellite images, a high correlation exists between neighboring image points which tend to occupy dense and restricted regions of the feature space. The image is divided into windows of the same size where the clustering is made. The classes obtained in several neighboring windows are clustered, and then again successively clustered until only one region corresponding to the whole image is obtained. By employing this algorithm only a few points are considered in each clustering, thus reducing computational effort. The method is illustrated as applied to LANDSAT images.

Dejesusparada, N.↗
NASA NTRS1984

A unified N-body and statistical treatment of stellar dynamics. I - The hybrid code. II - Applications to globular cluster cores

A unified N-body and statistical treatment of stellar dynamics is developed and applied to the late stages of core collapse and early stages of post collapse evolution in globular clusters. A 'hybrid' computer code is joined to a direct N-body code which is used to calculate exactly the behavior of particles in the inner spatial region, and the combination is used to follow particles statistically in the outer spatial region. A transition zone allows the exchange of particles and energy between the two regions. The main application results include: formation of a hard central binary system, reversal of core collapse and expansion due to the heat input from this binary, ejection of the binary from the core, and recollapse of the core; density profiles that form a one-parameter sequence during the core oscillations; and indications that these oscillations will eventually cease.

Mcmillan, S. L. W.↗
NASA NTRS1995

Analysis of developing laminar flows in circular pipes using a higher-order finite-difference technique

A higher-order finite-difference technique is developed to calculate the developing-flow field of steady incompressible laminar flows in the entrance regions of circular pipes. Navier-Stokes equations governing the motion of such a flow field are solved by using this new finite-difference scheme. This new technique can increase the accuracy of the finite-difference approximation, while also providing the option of using unevenly spaced clustered nodes for computation such that relatively fine grids can be adopted for regions with large velocity gradients. The velocity profile at the entrance of the pipe is assumed to be uniform for the computation. The velocity distribution and the surface pressure drop of the developing flow then are calculated and compared to existing experimental measurements reported in the literature. Computational results obtained are found to be in good agreement with existing experimental correlations and therefore, the reliability of the new technique has been successfully tested.

Gladden, Herbert J.↗
NASA NTRS1994

High Performance Distributed Computing in a Supercomputer Environment: Computational Services and Applications Issues

This tutorial proposes to be a practical guide for the uninitiated to the main topics and themes of high-performance computing (HPC), with particular emphasis to distributed computing. The intent is first to provide some guidance and directions in the rapidly increasing field of scientific computing using both massively parallel and traditional supercomputers. Because of their considerable potential computational power, loosely or tightly coupled clusters of workstations are increasingly considered as a third alternative to both the more conventional supercomputers based on a small number of powerful vector processors, as well as high massively parallel processors. Even though many research issues concerning the effective use of workstation clusters and their integration into a large scale production facility are still unresolved, such clusters are already used for production computing. In this tutorial we will utilize the unique experience made at the NAS facility at NASA Ames Research Center. Over the last five years at NAS massively parallel supercomputers such as the Connection Machines CM-2 and CM-5 from Thinking Machines Corporation and the iPSC/860 (Touchstone Gamma Machine) and Paragon Machines from Intel were used in a production supercomputer center alongside with traditional vector supercomputers such as the Cray Y-MP and C90.

Kramer, Williams T. C.↗
NASA NTRS2003

Glenn-ht/bem Conjugate Heat Transfer Solver for Large-scale Turbomachinery Models

A coupled Boundary Element/Finite Volume Method temperature-forward/flux-hack algorithm is developed for conjugate heat transfer (CHT) applications. A loosely coupled strategy is adopted with each field solution providing boundary conditions for the other in an iteration seeking continuity of temperature and heat flux at the fluid-solid interface. The NASA Glenn Navier-Stokes code Glenn-HT is coupled to a 3-D BEM steady state heat conduction code developed at the University of Central Florida. Results from CHT simulation of a 3-D film-cooled blade section are presented and compared with those computed by a two-temperature approach. Also presented are current developments of an iterative domain decomposition strategy accommodating large numbers of unknowns in the BEM. The blade is artificially sub-sectioned in the span-wise direction, 3-D BEM solutions are obtained in the subdomains, and interface temperatures are averaged symmetrically when the flux is updated while the fluxes are averaged anti-symmetrically to maintain continuity of heat flux when the temperatures are updated. An initial guess for interface temperatures uses a physically-based 1-D conduction argument to provide an effective starting point and significantly reduce iteration. 2-D and 3-D results show the process converges efficiently and offers substantial computational and storage savings. Future developments include a parallel multi-grid implementation of the approach under MPI for computation on PC clusters.

Divo, E.↗
NASA NTRS2004

What Multilevel Parallel Programs do when you are not Watching: A Performance Analysis Case Study Comparing MPI/OpenMP, MLP, and Nested OpenMP

With the current trend in parallel computer architectures towards clusters of shared memory symmetric multi-processors, parallel programming techniques have evolved that support parallelism beyond a single level. When comparing the performance of applications based on different programming paradigms, it is important to differentiate between the influence of the programming model itself and other factors, such as implementation specific behavior of the operating system (OS) or architectural issues. Rewriting-a large scientific application in order to employ a new programming paradigms is usually a time consuming and error prone task. Before embarking on such an endeavor it is important to determine that there is really a gain that would not be possible with the current implementation. A detailed performance analysis is crucial to clarify these issues. The multilevel programming paradigms considered in this study are hybrid MPI/OpenMP, MLP, and nested OpenMP. The hybrid MPI/OpenMP approach is based on using MPI [7] for the coarse grained parallelization and OpenMP [9] for fine grained loop level parallelism. The MPI programming paradigm assumes a private address space for each process. Data is transferred by explicitly exchanging messages via calls to the MPI library. This model was originally designed for distributed memory architectures but is also suitable for shared memory systems. The second paradigm under consideration is MLP which was developed by Taft. The approach is similar to MPi/OpenMP, using a mix of coarse grain process level parallelization and loop level OpenMP parallelization. As it is the case with MPI, a private address space is assumed for each process. The MLP approach was developed for ccNUMA architectures and explicitly takes advantage of the availability of shared memory. A shared memory arena which is accessible by all processes is required. Communication is done by reading from and writing to the shared memory.

Jost, Gabriele↗
NASA NTRS2015

Earth Science Data Fusion with Event Building Approach

Objectives of the NASA Information And Data System (NAIADS) project are to develop a prototype of a conceptually new middleware framework to modernize and significantly improve efficiency of the Earth Science data fusion, big data processing and analytics. The key components of the NAIADS include: Service Oriented Architecture (SOA) multi-lingual framework, multi-sensor coincident data Predictor, fast into-memory data Staging, multi-sensor data-Event Builder, complete data-Event streaming (a work flow with minimized IO), on-line data processing control and analytics services. The NAIADS project is leveraging CLARA framework, developed in Jefferson Lab, and integrated with the ZeroMQ messaging library. The science services are prototyped and incorporated into the system. Merging the SCIAMACHY Level-1 observations and MODIS/Terra Level-2 (Clouds and Aerosols) data products, and ECMWF re- analysis will be used for NAIADS demonstration and performance tests in compute Cloud and Cluster environments.

Lukashin, C.↗
NASA NTRS1994

X-ray clusters from a high-resolution hydrodynamic PPM simulation of the cold dark matter universe

A new three-dimensional hydrodynamic code based on the piecewise parabolic method (PPM) is utilized to compute the distribution of hot gas in the standard Cosmic Background Explorer (COBE)-normalized cold dark matter (CDM) universe. Utilizing periodic boundary conditions, a box with size 85 h(exp-1) Mpc, having cell size 0.31 h(exp-1) Mpc, is followed in a simulation with 270(exp 3)=10(exp 7.3) cells. Adopting standard parameters determined from COBE and light-element nucleosynthesis, Sigma(sub 8)=1.05, Omega(sub b)=0.06, we find the X-ray-emitting clusters, compute the luminosity function at several wavelengths, the temperature distribution, and estimated sizes, as well as the evolution of these quantities with redshift. The results, which are compared with those obtained in the preceding paper (Kang et al. 1994a), may be used in conjuction with ROSAT and other observational data sets. Overall, the results of the two computations are qualitatively very similar with regard to the trends of cluster properties, i.e., how the number density, radius, and temeprature depend on luminosity and redshift. The total luminosity from clusters is approximately a factor of 2 higher using the PPM code (as compared to the 'total variation diminishing' (TVD) code used in the previous paper) with the number of bright clusters higher by a similar factor. The primary conclusions of the prior paper, with regard to the power spectrum of the primeval density perturbations, are strengthened: the standard CDM model, normalized to the COBE microwave detection, predicts too many bright X-ray emitting clusters, by a factor probably in excess of 5. The comparison between observations and theoretical predictions for the evolution of cluster properties, luminosity functions, and size and temperature distributions should provide an important discriminator among competing scenarios for the development of structure in the universe.

Bryan, Greg L.↗
NASA NTRS2007

Radio Sources Toward Galaxy Clusters at 30 GHz

Extra-galactic radio sources are a significant contaminant in cosmic microwave background and Sunyaev-Zeldovich effect experiments. Deep interferometric observations with the BIMA and OVRO arrays are used to characterize the spatial, spectral, and flux distributions of radio sources toward massive galaxy clusters at 28.5 GHz. We compute counts of mJy source fluxes from 89 fields centered on known massive galaxy clusters and 8 non-cluster fields. We find that source counts in the inner regions of the cluster fields (within 0.5 arcmin of the cluster center) are a factor of 8.9 (+4.2 to -3.8) times higher than counts in the outer regions of the cluster fields (radius greater than 0.5 arcmin). Counts in the outer regions of the cluster fields are in turn a factor of 3.3 (+4.1 -1.8) greater than those in the noncluster fields. Counts in the non-cluster fields are consistent with extrapolations from the results of other surveys. We compute spectral indices of mJy sources in cluster fields between 1.4 and 28.5 GHz and find a mean spectral index of al[ja = 0.66 with an rms dispersion of 0.36, where flux S varies as upsilon(sup -alpha). The distribution is skewed, with a median spectral index of 0.72 and 25th and 75th percentiles of 0.51 and 0.92, respectively. This is steeper than the spectral indices of stronger field sources measured by other surveys.

Coble, K.↗
NASA NTRS1979

On the clustering of multidimensional pictorial data

Obvious approaches to reducing the cost (in computer resources) of applying current clustering techniques to the problem of remote sensing are discussed. The use of spatial information in finding fields and in classifying mixture pixels is examined, and the AMOEBA clustering program is described. Internally, a pattern recognition program, from without, AMOEBA appears to be an unsupervised clustering program. It is fast and automatic. No choices (such as arbitrary thresholds to set split/combine sequences) need be made. The problem of finding the number of clusters is solved automatically. At the conclusion of the program, all points in the scene are classified; however, a provision is included for a reject classification of some points which, within the theoretical framework, cannot rationally be assigned to any cluster.

Bryant, J. D.↗
NASA NTRS1972

Computer program documentation: ISOCLS iterative self-organizing clustering program, program C094

The author has identified the following significant results. This program implements an algorithm which, ideally, sorts a given set of multivariate data points into similar groups or clusters. The program is intended for use in the evaluation of multispectral scanner data; however, the algorithm could be used for other data types as well. The user may specify a set of initial estimated cluster means to begin the procedure, or he may begin with the assumption that all the data belongs to one cluster. The procedure is initiatized by assigning each data point to the nearest (in absolute distance) cluster mean. If no initial cluster means were input, all of the data is assigned to cluster 1. The means and standard deviations are calculated for each cluster.

Minter, R. T.↗