We address the problem of partitioning and dynamic load balancing on clusters with heterogeneous hardware resources. We propose DRUM, a model that encapsulates hardware resources and their interconnection topology. DRUM provides monitoring facilities for dynamic evaluation of communication, memory, and processing capabilities. Heterogeneity is quantified by merging the information from the monitors to produce a scalar number called 'power.' This power allows DRUM to be used easily by existing load-balancing procedures such as those in the Zoltan Toolkit while placing minimal burden on application programmers. We demonstrate the use of DRUM to guide load balancing in the adaptive solution of a Laplace equation on a heterogeneous cluster. We observed a significant reduction in execution time compared to traditional methods.
Many problems from engineering and the sciences require the solution of sequences of linear systems where the matrix and right-hand side change from one system to the next, and the linear systems are not available simultaneously. We review a class of Krylov subspace methods for sequences of linear systems, which can significantly reduce the cost of solving the next system in the sequence by 'recycling' subspace information from previous systems. These methods have been successfully applied to sequences of linear systems arising from several different application areas. We analyze a particular method, GCRO-DR, that recycles approximate invariant subspaces, and establish residual bounds that suggest a convergence rate similar to one obtained by removing select eigenvector components from the initial residual. We review implications of this analysis, which suggests problem classes where we expect this technique to be particularly effective. From this analysis and related numerical experiments we also demonstrate that recycling the invariant subspace corresponding to the eigenvalues of smallest absolute magnitude is often not the best choice, especially for nonsymmetric problems, and that GCRO-DR will, in practice, select better subspaces. These results suggest possibilities for improvement in the subspace selection process.
The present ITSE journal special issue on 'Learning About Social Interaction through Gaming' is the result of an invitation to the attendees of a one-day workshop on 'Social Learning Through Gaming' co-organized by the guest editors and held at the Human Factors in Computing Systems (CHI) conference on April 26, 2004 in Vienna, Austria. CHI is one of the premiere conferences on human-computer interaction. CHI 2004 attracted hundreds of delegates from all over the world. The CHI workshop program results from a competitive selection process. The Social Learning through Gaming workshop was filled to capacity and attended by approximately 25 participants from Europe and North America who submitted position papers that were refereed and selected for participation based on the relevancy and innovativeness of the research. The participants came together to share research on play, learning, games, interactive technologies, and what playing and designing games can teach us about social behaviors. The present special issue focuses on learning about social aspects through gaming: learning to socialize through games and learning games through social behavior.
The presence of mechanical joints--typified by the lap joint--in otherwise linear structures has been accommodated in structural dynamics via ad hoc methods for a century. The methods range from tuning linear models to approximate non-linear behavior in restricted load ranges to various methods which introduce joint dissipation in a post-processing stage. Other methods, employing constitutive models for the joints are being developed and their routine use is on the horizon.
We have measured the temperature dependence of mechanical dissipation in tetrahedral amorphous carbon flexural and torsional resonators over the temperature range from 300 to 1023 K. The mechanical dissipation was found to be controlled by defects within the material, and the magnitude and temperature dependence of the dissipation were found to depend on whether flexural or torsional vibrational modes were excited. The defects that were active under flexural stresses have a relatively flat concentration from 0.4 to 0.7 eV with an ever increasing defect concentration up to 1.9 eV. Under shear stresses (torsion), the defect activation energies increase immediately beginning at 0.4 eV, with increasing defect concentration at higher energies.
Solid solutions of lead-based perovskites are the backbone materials of the piezoelectric components for transducer, actuator, and resonator applications. These components, typically small in size, are fabricated from large sintered ceramic slugs using grinding and lapping processes. These operations increase manufacturing costs and produce a large hazardous waste stream, especially when component size decreases. To reduce costs and hazardous wastes associated with the production of these components, an injection molding technique is being investigated to replace the machining processes. The first step in the new technique is to compound an organic carrier with a ceramic powder. The organic carrier is a thermoplastic based system composed of a main carrier, a binder, and a surfactant. Understanding the rheology of the compounded material is necessary to minimize the creation of defects such as voids or cavities during the injection-molding process. An experiment was performed to model the effects of changes in the composition and processing of the material on the rheological behavior. Factors studied included: the surfactant of the organic carrier system, the solid loading of the compounded material, and compounding time. The effects of these factors on the viscosity of the material were investigated.
We report the experimental realization of a new type of optical parametric oscillator in which oscillation is achieved by polarization rotation in a linear retarder, followed by nonlinear polarization mixing. The mixing is performed by a type II degenerate parametric downconversion in a periodically poled KTP crystal pumped by a 1064 nm pulsed Nd:YAG pump. A single, linearly polarized beam, precisely at the degenerate wavelength is generated. The output spectrum has a narrow linewidth (below the instrumentation bandwidth of 1 nm) and is highly stable with respect to variations in the crystal temperature.
Concepts from Complexity Science are valuable and allow a simulation approach for critical infrastructures that is flexible and has wide ranging applications.
A large-scale field demonstration comparing final landfill cover designs was constructed and monitored at Sandia National Laboratories in Albuquerque, New Mexico. Two conventional designs (a RCRA Subtitle 'D' Soil Cover and a RCRA Subtitle 'C' Compacted Clay Cover) were constructed side-by-side with four alternative cover test plots designed for arid environments. The demonstration was intended to evaluate the various cover designs based on their respective water balance performance, ease and reliability of construction, and cost. A portion of this project involves the characterization of vegetation establishment and growth on the landfill covers. The various prototype landfill covers were expected to have varying flux rates (Dwyer et al 2000). The landfill covers were further expected to influence vegetation establishment and growth, which may impact site erosion potential and long-term site integrity. Objectives of this phase were to quantify the types of plants occupying each site, the percentage of ground covered by these plants, the density (number of plants per unit area) of plants, and the plant biomass production. The results of this vegetation analysis are presented in this report.3 DRAFT07/06/14AcknowledgementsWe would like to thank all technical and support staff from Sandia and the USDA Forest Service's Rocky Mountain Station not included in the authors' list of this document for their valuable contributions to this research. We would also like to acknowledge the Department of Energy's Subsurface Contaminants Focus Area for funding this work.4
This Work Plan identifies and outlines interim measures to address nitrate contamination in groundwater at the Burn Site, Sandia National Laboratories/New Mexico. The New Mexico Environment Department has required implementation of interim measures for nitrate-contaminated groundwater at the Burn Site. The purpose of interim measures is to prevent human or environmental exposure to nitrate-contaminated groundwater originating from the Burn Site. This Work Plan details a summary of current information about the Burn Site, interim measures activities for stabilization, and project management responsibilities to accomplish this purpose.
We show that the orientation of pentacene molecules is controlled by the electronic structure of the surface on which they are deposited. We suggest that the near-Fermi level density of states above the surface controls the interaction of the substrate with the pentacene {pi} orbitals. A reduction of this density as compared to noble metals, realized in semimetallic Bi(001) and Si(111)(5 x 2)Au surfaces, results in pentacene standing up. Interestingly, pentacene grown on Bi(001) is highly ordered, yielding the first vertically oriented epitaxial pentacene thin films observed to date.
This report documents activities related to the ASCI AD Resistance Weld Process Modeling Project AD2003-15. Activities up to and including FY2004 are discussed. This was the third year for this multi year project, the objective of which is to position the SIERRA computational tools for the solution of resistance welding problems. The process of interest is a three-way coupled problem involving current flow, temperature buildup and large plastic deformation. The DSW application is the reclamation stem weld used in the manufacture of high pressure gas bottles. This is the first year the CALAGIO suite of codes (eCALORE, CALORE, and ADAGIO) was used to successfully solve a three-way coupled problem in SIERRA. This report discusses the application of CALAGIO to the tapered bar acceptance problem and a similar but independent tapered bar simulation of a companion C6 experiment. New additions to the EMMI constitutive model and issues related to CALAGIO performance are also discussed.
Classical density functional theory (DFT) is applied to study properties of fully detailed, realistic models of poly(dimethylsiloxane) liquids near silica surfaces and compared to results from molecular dynamics simulations. In solving the DFT equations, the direct correlation functions are obtained from the polymer reference interaction site model (PRISM) theory for the repulsive parts of the interatomic interactions, and the attractions are treated via the random-phase approximation (RPA). Good agreement between density profiles calculated from DFT and from the simulations is obtained with empirical scaling of the direct correlation functions. Separate scaling factors are required for the PRISM and RPA parts of the direct correlation functions. Theoretical predictions of stress profiles, normal pressure, and surface tensions are also in reasonable agreement with simulation results.
In the crystal structure of the title compound, C{sub 4}H{sub 4}N{sub 2}O{sub 3}, the packing is dominated by intermolecular carbonyl-carbonyl interactions and N-H...O hydrogen bonds.
The solution of the governing steady transport equations for momentum, heat and mass transfer in fluids undergoing non-equilibrium chemical reactions can be extremely challenging. The difficulties arise from both the complexity of the nonlinear solution behavior as well as the nonlinear, coupled, non-symmetric nature of the system of algebraic equations that results from spatial discretization of the PDEs. In this paper, we briefly review progress on developing a stabilized finite element (FE) capability for numerical solution of these challenging problems. The discussion considers the stabilized FE formulation for the low Mach number Navier-Stokes equations with heat and mass transport with non-equilibrium chemical reactions, and the solution methods necessary for detailed analysis of these complex systems. The solution algorithms include robust nonlinear and linear solution schemes, parameter continuation methods, and linear stability analysis techniques. Our discussion considers computational efficiency, scalability, and some implementation issues of the solution methods. Computational results are presented for a CFD benchmark problem as well as for a number of large-scale, 2D and 3D, engineering transport/reaction applications.