Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Grid Partition”

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 127 records · Page 7

Distributing Power Grid State Estimation on HPC Clusters A System Architecture Prototype

The future power grid is expected to further expand with highly distributed energy sources and smart loads. The increased size and complexity lead to increased burden on existing computational resources in energy control centers. Thus the need to perform real-time assessment on such systems entails efficient means to distribute centralized functions such as state estimation in the power system. In this paper, we present our early prototype of a system architecture that connects distributed state estimators individually running parallel programs to solve non-linear estimation procedure. The prototype consists of a middleware and data processing toolkits that allows data exchange in the distributed state estimation. We build a test case based on the IEEE 118 bus system and partition the state estimation of the whole system model to available HPC clusters. The measurement from the testbed demonstrates the low overhead of our solution.

Distributed architecture↗

Reduced order modelling for dynamic simulations: LDRD feasibility study final report

This feasibility study (12-FS-007) investigated the current state-of-the-art for the construction of reduced order models (ROMs) from dynamic PDE-based simulations. One novel result of was the demonstration of a time-partitioned ROM for the advection-diffusion equation that achieved high accuracy with a small ROM size. The time-domain of a model PDE was partitioned and a fixed-size ROM was tuned to each subdomain resulting in an effective method for achieving an efficient and accurate ROM for a dynamic problem. This LDRD also investigated the construction ROMs for simulations that utilize overlapping grids. In this case, it was found that weighting the discrete inner product in a manner that approximates the continuous inner product yields a significantly more accurate ROM. This project demonstrated the feasibility of applying reduced order models to complex PDE-based simulation tools and explored the benefits of adaptive ROM techniques. The results of the work led to an Exploratory Research LDRD, funded for FY13, that will focus on the development of composite adaptive reduced order models suitable for use with a number of LLNL applications.

97 MATHEMATICS AND COMPUTING↗

Accelerating Computation of the Unit Commitment Problem (Presentation)

Production cost models (PCMs) simulate power system operation at hourly (or higher) resolution. While computation times often extend into multiple days, the sequential nature of PCM's makes parallelism difficult. We exploit the persistence of unit commitment decisions to select partition boundaries for simulation horizon decomposition and parallel computation. Partitioned simulations are benchmarked against sequential solutions for optimality and computation time.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Modeling pre-Exascale AMR Parallel I/O Workloads via Proxy Applications

The present work investigates the modeling of preexascale input/output (I/O) workloads of Adaptive Mesh Refinement (AMR) simulations through a simple proxy application. We collect data from the AMReX Castro framework running on the Summit supercomputer for a wide range of scales and mesh partitions for the hydrodynamic Sedov case as a baseline to provide sufficient coverage to the formulated proxy model. The non-linear analysis data production rates are quantified as a function of a set of input parameters such as output frequency, grid size, number of levels, and the Courant-Friedrichs-Lewy (CFL) condition number for each rank, mesh level and simulation time step. Linear regression is then applied to formulate a simple analytical model which allows to translate AMReX inputs into MACSio proxy I/O application parameters, resulting in a simple “kernel” approximation for data production at each time step. Results show that MACSio can simulate actual AMReX nonlinear “static” I/O workloads to a certain degree of confidence on the Summit supercomputer using the present methodology. The goal is to provide an initial level of understanding of AMR I/O workloads via lightweight proxy applications models to facilitate autotune data management strategies in anticipation of exascale systems.

Godoy, William↗

A stable partitioned FSI algorithm for incompressible flow and deforming beams

An added-mass partitioned (AMP) algorithm is described for solving fluid–structure interaction (FSI) problems coupling incompressible flows with thin elastic structures undergoing finite deformations. The new AMP scheme is fully second-order accurate and stable, without sub-time-step iterations, even for very light structures when added-mass effects are strong. The fluid, governed by the incompressible Navier–Stokes equations, is solved in velocity-pressure form using a fractional-step method; large deformations are treated with a mixed Eulerian-Lagrangian approach on deforming composite grids. The motion of the thin structure is governed by a generalized Euler–Bernoulli beam model, and these equations are solved in a Lagrangian frame using two approaches, one based on finite differences and the other on finite elements. The key AMP interface condition is a generalized Robin (mixed) condition on the fluid pressure. This condition, which is derived at a continuous level, has no adjustable parameters and is applied at the discrete level to couple the partitioned domain solvers. Special treatment of the AMP condition is required to couple the finite-element beam solver with the finite-difference-based fluid solver, and two coupling approaches are described. A normal-mode stability analysis is performed for a linearized model problem involving a beam separating two fluid domains, and it is shown that the AMP scheme is stable independent of the ratio of the mass of the fluid to that of the structure. A traditional partitioned (TP) scheme using a Dirichlet–Neumann coupling for the same model problem is shown to be unconditionally unstable if the added mass of the fluid is too large. A series of benchmark problems of increasing complexity are considered to illustrate the behavior of the AMP algorithm, and to compare the behavior with that of the TP scheme. The results of all these benchmark problems verify the stability and accuracy of the AMP scheme. Results for one benchmark problem modeling blood flow in a deforming artery are also compared with corresponding results available in the literature.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

The Effects of Disturbance and Climate on Carbon Storage and the Exchanges of CO 2 Water Vapor and Energy Exchange of Evergreen Coniferous Forests in the Pacific Northwest: Integration of Eddy Flux, Plant and Soil Measurements at a Cluster of Supersites. Final report

This is the final technical report containing a summary of all findings with regard to the following objectives of the project: (1) To quantify and understand the effects of wildfire on carbon storage and the exchanges of energy, CO2, and water vapor in a chronosequence of ponderosa pine (disturbance gradient); (2) To investigate the effects of seasonal and interannual variation in climate on carbon storage and the exchanges of energy, CO2, and water vapor in mature conifer forests in two climate zones: mesic 40-yr old Douglas-fir and semi-arid 60-yr old ponderosa pine (climate gradient); (3) To reduce uncertainty in estimates of CO2 feedbacks to the atmosphere by providing an improved model formulation for existing biosphere-atmosphere models; and (4) To provide high quality data for AmeriFlux and the NACP on micrometeorology, meteorology, and biology of these systems. Objective (1): A study integrating satellite remote sensing, AmeriFlux data, and field surveys in a simulation modeling framework estimated that the pyrogenic carbon emissions, tree mortality, and net carbon exchange associated with four large wildfires that burned ~50,000 hectares in 2002-2003 were equivalent to 2.4% of Oregon statewide anthropogenic carbon emissions over the same two-year period. Most emissions were from the combustion of the forest floor and understory vegetation, and only about 1% of live tree mass was combusted on average. Objective (2): A study of multi-year flux records across a chronosequence of ponderosa pine forests yielded that the net carbon uptake is over three times greater at a mature pine forest compared with young pine. The larger leaf area and wetter and cooler soils of the mature forest mainly caused this effect. A study analyzing seven years of carbon and water dynamics showed that interannual and seasonal variability of net carbon exchange was primarily related to variability in growing season length, which was a linear function of plant-available soil moisture in spring and early summer. A multi-year drought (2001-2003) led to a significant reduction of net ecosystem exchange due to carry-over effects in soil moisture and carbohydrate reserves in plant-tissue. In the same forest, the interannual variability in the rate carbon is lost from the soil and forest floor is considerable and related to the variability in tree growth as much as it is to variability in soil climatic conditions. Objective (3): Flux data from the mature ponderosa pine site support a physical basis for filtering nighttime data with friction velocity above the canopy. An analysis of wind fields and heat transport in the subcanopy at the mesic 40-year old Douglas site yielded that the non-linear structure and behavior of spatial temperature gradients and the flow field require enhanced sensor networks to estimate advective fluxes in the subcanopy of forest to close the surface energy balance in forests. Reliable estimates for flux uncertainties are needed to improve model validation and data assimilation in process-based carbon models, inverse modeling studies and model-data synthesis, where the uncertainties may be as important as the fluxes themselves. An analysis of the time scale dependence of the random and flux sampling error yielded that the additional flux obtained by increasing the perturbation timescale beyond about 10 minutes is dominated by random sampling error, and therefore little confidence can be placed in its value. Artificial correlation between gross ecosystem productivity (GEP) and ecosystem respiration (Re) is a consequence of flux partitioning of eddy covariance flux data when GEP is computed as the difference between NEE and computed daytime Re (e.g. using nighttime Re extrapolated into daytime using soil or air temperatures). Tower-data must be adequately spatially averaged before comparison to gridded model output as the time variability of both is inherently different. The eddy-covariance data collected at the mature ponderosa pine site and the mesic Douglas fir site were used to develop and evaluate a new method to extract the signal of ecosystem respiration directly from daytime net ecosystem exchange. This approach may help reducing uncertainty in carbon budgets by providing direct measurements of ecosystem respiration during daylight conditions by replacing modeled estimates. Objective (4): We submitted our flux and biological and ancillary data to the AmeriFlux web site and to Fluxnet. This includes atmospheric carbon, water, and heat fluxes, soil fluxes, NPP, carbon stocks, LAI, and disturbance history. Fluxnet is updating the original La Thuile files and will include the more recent years of data. They will be using a new approach to compute GPP, following discussions within the network about the need to improve GPP methodology

54 ENVIRONMENTAL SCIENCES↗

Seasonal responses of terrestrial ecosystem water‐use efficiency to climate change

Abstract Ecosystem water‐use efficiency ( EWUE ) is an indicator of carbon–water interactions and is defined as the ratio of carbon assimilation ( GPP ) to evapotranspiration ( ET ). Previous research suggests an increasing long‐term trend in annual EWUE over many regions and is largely attributed to the physiological effects of rising CO 2 . The seasonal trends in EWUE , however, have not yet been analyzed. In this study, we investigate seasonal EWUE trends and responses to various drivers during 1982–2008. The seasonal cycle for two variants of EWUE , water‐use efficiency ( WUE , GPP / ET ), and transpiration‐based WUE ( WUE t , the ratio of GPP and transpiration), is analyzed from 0.5° gridded fields from four process‐based models and satellite‐based products, as well as a network of 63 local flux tower observations. WUE derived from flux tower observations shows moderate seasonal variation for most latitude bands, which is in agreement with satellite‐based products. In contrast, the seasonal EWUE trends are not well captured by the same satellite‐based products. Trend analysis, based on process‐model factorial simulations separating effects of climate, CO 2 , and nitrogen deposition ( NDEP ), further suggests that the seasonal EWUE trends are mainly associated with seasonal trends of climate, whereas CO 2 and NDEP do not show obvious seasonal difference in EWUE trends. About 66% grid cells show positive annual WUE trends, mainly over mid‐ and high northern latitudes. In these regions, spring climate change has amplified the effect of CO 2 in increasing WUE by more than 0.005 gC m −2 mm −1 yr −1 for 41% pixels. Multiple regression analysis further shows that the increase in springtime WUE in the northern hemisphere is the result of GPP increasing faster than ET because of the higher temperature sensitivity of GPP relative to ET . The partitioning of annual EWUE to seasonal components provides new insight into the relative sensitivities of GPP and ET to climate, CO 2, and NDEP .

Huang, Mengtian↗

Partition-of-unity finite-element method for large scale quantum molecular dynamics on massively parallel computational platforms

Over the course of the past two decades, quantum mechanical calculations have emerged as a key component of modern materials research. However, the solution of the required quantum mechanical equations is a formidable task and this has severely limited the range of materials systems which can be investigated by such accurate, quantum mechanical means. The current state of the art for large-scale quantum simulations is the planewave (PW) method, as implemented in now ubiquitous VASP, ABINIT, and QBox codes, among many others. However, since the PW method uses a global Fourier basis, with strictly uniform resolution at all points in space, and in which every basis function overlaps every other at every point, it suffers from substantial inefficiencies in calculations involving atoms with localized states, such as first-row and transition-metal atoms, and requires substantial nonlocal communications in parallel implementations, placing critical limits on scalability. In recent years, real-space methods such as finite-differences (FD) and finite-elements (FE) have been developed to address these deficiencies by reformulating the required quantum mechanical equations in a strictly local representation. However, while addressing both resolution and parallel-communications problems, such local real-space approaches have been plagued by one key disadvantage relative to planewaves: excessive degrees of freedom (grid points, basis functions) needed to achieve the required accuracies. And so, despite critical limitations, the PW method remains the standard today. In this work, we show for the first time that this key remaining disadvantage of real-space methods can in fact be overcome: by building known atomic physics into the solution process using modern partition-of-unity (PU) techniques in finite element analysis. Indeed, our results show order-of-magnitude reductions in basis size relative to state-of-the-art planewave based methods. The method developed here is completely general, applicable to any crystal symmetry and to both metals and insulators alike. We have developed and implemented a full self-consistent Kohn-Sham method, including both total energies and forces for molecular dynamics, and developed a full MPI parallel implementation for large-scale calculations. We have applied the method to the gamut of physical systems, from simple insulating systems with light atoms to complex d- and f-electron systems, requiring large numbers of atomic-orbital enrichments. In every case, the new PU FE method attained the required accuracies with substantially fewer degrees of freedom, typically by an order of magnitude or more, than the current state-of-the-art PW method. Finally, our initial MPI implementation has shown excellent parallel scaling of the most time-critical parts of the code up to 1728 processors, with clear indications of what will be required to achieve comparable scaling for the rest. Having shown that the key remaining disadvantage of real-space methods can in fact be overcome, the work has attracted significant attention: with sixteen invited talks, both domestic and international, so far; two papers published and another in preparation; and three new university and/or national laboratory collaborations, securing external funding to pursue a number of related research directions. Having demonstrated the proof of principle, work now centers on the necessary extensions and optimizations required to bring the prototype method and code delivered here to production applications.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Provably optimal parallel transport sweeps on regular grids

We have found provably optimal algorithms for full-domain discrete-ordinate transport sweeps on regular grids in 3D Cartesian geometry. We describe these algorithms and sketch a 'proof that they always execute the full eight-octant sweep in the minimum possible number of stages for a given P{sub x} x P{sub y} x P{sub z} partitioning. Computational results demonstrate that our optimal scheduling algorithms execute sweeps in the minimum possible stage count. Observed parallel efficiencies agree well with our performance model. An older version of our PDT transport code achieves almost 80% parallel efficiency on 131,072 cores, on a weak-scaling problem with only one energy group, 80 directions, and 4096 cells/core. A newer version is less efficient at present-we are still improving its implementation - but achieves almost 60% parallel efficiency on 393,216 cores. These results conclusively demonstrate that sweeps can perform with high efficiency on core counts approaching 10{sup 6}. (authors)

97 MATHEMATICS AND COMPUTING↗

Quasi-Optimal Elimination Trees for 2D Grids with Singularities

We construct quasi-optimal elimination trees for 2D finite element meshes with singularities. These trees minimize the complexity of the solution of the discrete system. The computational cost estimates of the elimination process model the execution of the multifrontal algorithms in serial and in parallel shared-memory executions. Since the meshes considered are a subspace of all possible mesh partitions, we call these minimizers quasi-optimal. We minimize the cost functionals using dynamic programming. Finding these minimizers is more computationally expensive than solving the original algebraic system. Nevertheless, from the insights provided by the analysis of the dynamic programming minima, we propose a heuristic construction of the elimination trees that has cost O N e log ⁡ N e , where N e is the number of elements in the mesh. We show that this heuristic ordering has similar computational cost to the quasi-optimal elimination trees found with dynamic programming and outperforms state-of-the-art alternatives in our numerical experiments.

Paszyńska, A.↗

High performance Data Driven Agent-based Modeling Framework for Simulation of Commute Mode Choices in Metropolitan Area

Active transportation, human-powered transportation modes such as walking and bicycling, not only reduces the carbon footprint from the transportation sector but also promotes healthy living by offering opportunities for people to build physical activity into their daily routine. To encourage active transportation through urban planning and public campaigns, it is of significant importance to infer factors that substantially influence commuters in their transportation mode choice process. This necessitates a flexible and repeatable tool that can evaluate how a policy is perceived by individual commuters and convert their decisions into macro level understanding. This paper introduces one such effort that is specifically designed for studies of transportation mode choices in metropolitan areas. It provides results from a high-resolution data driven simulation based on high performance computing implementation of the agent-based model framework for home-to-work commute trips. The framework uses a graph-partition based technique that can leverage the interaction structure of agents within a geographic proximity and can boost the simulation execution time. Further, based on a flexible design, it can run ABM with different levels of computing resources-from multi core workstations to an HPC grid. The framework has been tested on the Titan Cray XK7 supercomputer of the Oak Ridge Leadership Computing Facility.

Park, Byung↗

Functional approach to exploring climatic and landscape controls of runoff generation: 1. Behavioral constraints on runoff volume

Abstract Inspired by the Dunne diagram, the climatic and landscape controls on the partitioning of annual runoff into its various components (Hortonian and Dunne overland flow and subsurface stormflow) are assessed quantitatively, from a purely theoretical perspective. A simple distributed hydrologic model has been built sufficient to simulate the effects of different combinations of climate, soil, and topography on the runoff generation processes. The model is driven by a sequence of simple hypothetical precipitation events, for a large combination of climate and landscape properties, and hydrologic responses at the catchment scale are obtained through aggregation of grid‐scale responses. It is found, first, that the water balance responses, including relative contributions of different runoff generation mechanisms, could be related to a small set of dimensionless similarity parameters. These capture the competition between the wetting, drying, storage, and drainage functions underlying the catchment responses, and in this way, provide a quantitative approximation of the conceptual Dunne diagram. Second, only a subset of all hypothetical catchment/climate combinations is found to be “behavioral,” in terms of falling sufficiently close to the Budyko curve, describing mean annual runoff as a function of climate aridity. Furthermore, these behavioral combinations are mostly consistent with the qualitative picture presented in the Dunne diagram, indicating clearly the commonality between the Budyko curve and the Dunne diagram. These analyses also suggest clear interrelationships amongst the “behavioral” climate, soil, and topography parameter combinations, implying these catchment properties may be constrained to be codependent in order to satisfy the Budyko curve.

Li, Hong‐Yi↗

Variations of characteristic time scales in rotating stratified turbulence using a large parametric numerical study

In this work, we study rotating stratified turbulence (RST) making use of numerical data stemming from a large parametric study varying the Reynolds, Froude and Rossby numbers, Re, Fr and Ro in a broad range of values. The computations are performed using periodic boundary conditions on grids of 10243 points, with no modeling of the small scales, no forcing and with large-scale random initial conditions for the velocity field only, and there are altogether 65 runs analyzed in this paper. The buoyancy Reynolds number defined as R B = ReFr 2 varies from negligible values to ≈ 105, approaching atmospheric or oceanic regimes. This preliminary analysis deals with the variation of characteristic time scales of RST with dimensionless parameters, focusing on the role played by the partition of energy between the kinetic and potential modes, as a key ingredient for modeling the dynamics of such flows. We find that neither rotation nor the ratio of the Brunt-Väisälä frequency to the inertial frequency seem to play a major role in the absence of forcing in the global dynamics of the small-scale kinetic and potential modes. Specifically, in these computations, mostly in regimes of wave turbulence, characteristic times based on the ratio of energy to dissipation of the velocity and temperature fluctuations, T V and T P , vary substantially with parameters. Their ratio γ=T V /T P follows roughly a bell-shaped curve in terms of Richardson number Ri. It reaches a plateau --on which time scales become comparable, γ≈0.6 -- when the turbulence has significantly strengthened, leading to numerous destabilization events together with a tendency towards an isotropization of the flow.

97 MATHEMATICS AND COMPUTING↗

Architecture for the Next Generation System Management Tools

To get more results or greater accuracy, computational scientists execute their applications on distributed computing platforms such as Clusters, Grids and Clouds. These platforms are different in terms of hardware and software resources as well as locality: some span across multiple sites and multiple administrative domains whereas others are limited to a single site/domain. As a consequence, in order to scale their applica- tions up the scientists have to manage technical details for each target platform. From our point of view, this complexity should be hidden from the scientists who, in most cases, would prefer to focus on their research rather than spending time dealing with platform configuration concerns. In this article, we advocate for a system management framework that aims to automatically setup the whole run-time environment according to the applications needs. The main difference with regards to usual approaches is that they generally only focus on the software layer whereas we address both the hardware and the software expecta- tions through a unique system. For each application, scientists describe their requirements through the definition of a Virtual Platform (VP) and a Virtual System Environment (VSE). Relying on the VP/VSE definitions, the framework is in charge of: (i) the configuration of the physical infrastructure to satisfy the VP requirements, (ii) the setup of the VP, and (iii) the customization of the execution environment (VSE) upon the former VP. We propose a new formalism that the system can rely upon to successfully perform each of these three steps without burdening the user with the specifics of the configuration for the physical resources, and system management tools. This formalism leverages Goldberg s theory for recursive virtual machines by introducing new concepts based on system virtualization (identity, partitioning, aggregation) and emulation (simple, abstraction). This enables the definition of complex VP/VSE configurations without making assumptions about the hardware and the software re- sources. For each requirement, the system executes the corresponding operation with the appropriate management tool. As a proof of concept, we implemented a first prototype that currently interacts with several system management tools (e.g., OSCAR, the Grid 5000 toolkit, and XtreemOS) and that can be easily extended to integrate new resource brokers or cloud systems such as Nimbus, OpenNebula or Eucalyptus for instance.

99 GENERAL AND MISCELLANEOUS↗

On the energy footprint of I/O management in Exascale HPC systems

The advent of unprecedentedly scalable yet energy hungry Exascale supercomputers poses a major challenge in sustaining a high performance-per-watt ratio. With I/O management acquiring a crucial role in supporting scientific simulations, various I/O management approaches have been proposed to achieve high performance and scalability. But, the details of how these approaches affect energy consumption have not been studied yet. Therefore, this paper aims to explore how much energy a supercomputer consumes while running scientific simulations when adopting various I/O management approaches. In particular, we closely examine three radically different I/O schemes including time partitioning, dedicated cores, and dedicated nodes. To accomplish this, we implement the three approaches within the Damaris I/O middleware and perform extensive experiments with one of the target HPC applications of the Blue Waters sustained-petaflop supercomputer project: the CM1 atmospheric model. Our experimental results obtained on the French Grid'5000 platform highlight the differences among these three approaches and illustrate in which way various configurations of the application and of the system can impact performance and energy consumption. Moreover, we propose and validate a mathematical model that estimates the energy consumption of a HPC simulation under different I/O approaches. This proposed model gives hints to pre-select the most energy-efficient I/O approach for a particular simulation on a particular HPC system and therefore provides a step towards energy-efficient HPC simulations in Exascale systems. To the best of our knowledge, our work provides the first in-depth look into the energy-performance tradeoffs of I/O management approaches.

97 MATHEMATICS AND COMPUTING↗

Learning Provably Stable Local Volt/Var Controllers for Efficient Network Operation

Here this paper develops a data-driven framework to synthesize local Volt/Var control strategies for distributed energy resources (DERs) in power distribution grids (DGs). Aiming to improve DG operational efficiency, as quantified by a generic optimal reactive power flow (ORPF) problem, we propose a two-stage approach. The first stage involves learning the manifold of optimal operating points determined by an ORPF instance. To synthesize local Volt/Var controllers, the learning task is partitioned into learning local surrogates (one per DER) of the optimal manifold with voltage input and reactive power output. Since these surrogates characterize efficient DG operating points, in the second stage, we develop local control schemes that steer the DG to these operating points. We identify the conditions on the surrogates and control parameters to ensure that the locally acting controllers collectively converge, in a global asymptotic sense, to a DG operating point agreeing with the local surrogates. We use neural networks to model the surrogates and enforce the identified conditions in the training phase. AC power flow simulations on the IEEE 37-bus network empirically bolster the theoretical stability guarantees obtained under linearized power flow assumptions. The tests further highlight the optimality improvement compared to prevalent benchmark methods.

24 POWER TRANSMISSION AND DISTRIBUTION↗