Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Distributed computation”

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 451 records · Page 25
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

A novel resource sharing algorithm based on distributed construction for radiant enclosure problems

This study demonstrates a novel approach to solving inverse radiant enclosure problems based on distributed construction. Specifically, the problem of determining the temperature distribution needed on the heater surfaces to achieve a desired design surface temperature profile is recast as a distributed construction problem in which a shared resource, temperature, is distributed by computational agents moving blocks. The sharing of blocks between agents enables them to achieve their desired local state, which in turn achieves the desired global state. Each agent uses the current state of their local environment and a simple set of rules to determine when to exchange blocks, each block representing a discrete unit of temperature change. This algorithm is demonstrated using the established two-dimensional inverse radiation enclosure problem. The temperature profile on the heater surfaces is adjusted to achieve a desired temperature profile on the design surfaces. The resource sharing algorithm was able to determine the needed temperatures on the heater surfaces to obtain the desired temperature distribution on the design surfaces in the nine cases examined.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Flexible silicon photonic architecture for accelerating distributed deep learning

The increasing size and complexity of deep learning (DL) models have led to the wide adoption of distributed training methods in datacenters (DCs) and high-performance computing (HPC) systems. However, communication among distributed computing units (CUs) has emerged as a major bottleneck in the training process. In this study, we propose Flex-SiPAC, a flexible silicon photonic accelerated compute cluster designed to accelerate multi-tenant distributed DL training workloads. Flex-SiPAC takes a co-design approach that combines a silicon photonic hardware platform with a tailored collective algorithm, optimized to leverage the unique physical properties of the architecture. The hardware platform integrates a novel wavelength-reconfigurable transceiver design and a micro-resonator-based wavelength-reconfigurable switch, enabling the system to achieve flexible bandwidth steering in the wavelength domain. The collective algorithm is designed to support reconfigurable topologies, enabling efficient all-reduce communications that are commonly used in DL training. The feasibility of the Flex-SiPAC architecture is demonstrated through two testbed experiments. First, an optical testbed experiment demonstrates the flexible routing of wavelengths by shuffling an array of input wavelengths using a custom-designed spatial-wavelength selective switch. Second, a four-GPU testbed running two DL workloads shows a 23% improvement in job completion time compared to a similarly sized leaf-spine topology. We further evaluate Flex-SiPAC using large-scale simulations, which show that Flex-SiPAC is able to reduce the communication time by 26% to 29% compared to state-of-the-art compute clusters under representative collective operations.

Wu, Zhenguo (ORCID:0000000322847985)↗
DOE OSTI2024

How to Build a Quantum Supercomputer: Scaling from Hundreds to Millions of Qubits

In the span of four decades, quantum computation has evolved from an intellectual curiosity to a potentially realizable technology. Today, small-scale demonstrations have become possible for quantum algorithmic primitives on hundreds of physical qubits and proof-of-principle error-correction on a single logical qubit. Nevertheless, despite significant progress and excitement, the path toward a full-stack scalable technology is largely unknown. There are significant outstanding quantum hardware, fabrication, software architecture, and algorithmic challenges that are either unresolved or overlooked. These issues could seriously undermine the arrival of utility-scale quantum computers for the foreseeable future. Here, we provide a comprehensive review of these scaling challenges. We show how the road to scaling could be paved by adopting existing semiconductor technology to build much higher-quality qubits, employing system engineering approaches, and performing distributed quantum computation within heterogeneous high-performance computing infrastructures. These opportunities for research and development could unlock certain promising applications, in particular, efficient quantum simulation/learning of quantum data generated by natural or engineered quantum systems. To estimate the true cost of such promises, we provide a detailed resource and sensitivity analysis for classically hard quantum chemistry calculations on surface-code error-corrected quantum computers given current, target, and desired hardware specifications based on superconducting qubits, accounting for a realistic distribution of errors. Furthermore, we argue that, to tackle industry-scale classical optimization and machine learning problems in a cost-effective manner, heterogeneous quantum-probabilistic computing with custom-designed accelerators should be considered as a complementary path toward scalability.

Mohseni, Masoud↗
DOE OSTI2015

Architectural Improvements and New Processing Tools for the Open XAL Online Model

The online model is the component of Open XAL providing accelerator modeling, simulation, and dynamic synchronization to live hardware. Significant architectural changes and feature additions have been recently made in two separate areas: 1) the managing and processing of simulation data, and 2) the modeling of RF cavities. Simulation data and data processing have been completely decoupled. A single class manages all simulation data while standard tools were developed for processing the simulation results. RF accelerating cavities are now modeled as composite structures where parameter and dynamics computations are distributed. The beam and hardware models both maintain their relative phase information, which allows for dynamic phase slip and elapsed time computation.

Accelerator Control↗
DOE OSTI2016

A Distributed OpenCL Framework using Redundant Computation and Data Replication

Applications written solely in OpenCL or CUDA cannot execute on a cluster as a whole. Most previous approaches that extend these programming models to clusters are based on a common idea: designating a centralized host node and coordinating the other nodes with the host for computation. However, the centralized host node is a serious performance bottleneck when the number of nodes is large. In this paper, we propose a scalable and distributed OpenCL framework called SnuCL-D for large-scale clusters. SnuCL-D's remote device virtualization provides an OpenCL application with an illusion that all compute devices in a cluster are confined in a single node. To reduce the amount of control-message and data communication between nodes, SnuCL-D replicates the OpenCL host program execution and data in each node. We also propose a new OpenCL host API function and a queueing optimization technique that significantly reduce the overhead incurred by the previous centralized approaches. To show the effectiveness of SnuCL-D, we evaluate SnuCL-D with a microbenchmark and eleven benchmark applications on a large-scale CPU cluster and a medium-scale GPU cluster.

Kim, Junghyun↗
DOE OSTI2017

AN INFORMATION-THEORETIC APPROACH TO OPTIMIZE JWST OBSERVATIONS AND RETRIEVALS OF TRANSITING EXOPLANET ATMOSPHERES

We provide an example of an analysis to explore the optimization of observations of transiting hot Jupiters with the James Webb Space Telescope ( JWST ) to characterize their atmospheres based on a simple three-parameter forward model. We construct expansive forward model sets for 11 hot Jupiters, 10 of which are relatively well characterized, exploring a range of parameters such as equilibrium temperature and metallicity, as well as considering host stars over a wide range in brightness. We compute posterior distributions of our model parameters for each planet with all of the available JWST spectroscopic modes and several programs of combined observations and compute their effectiveness using the metric of estimated mutual information per degree of freedom. From these simulations, clear trends emerge that provide guidelines for designing a JWST observing program. We demonstrate that these guidelines apply over a wide range of planet parameters and target brightnesses for our simple forward model.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2018

On the Usage of Ultrasound Computational Models for Decision Making Under Ambiguity

Computer modeling and simulation is becoming pervasive within the non-destructive evaluation (NDE) industry as a convenient tool for designing and assessing inspection techniques. This raises a pressing need for developing quantitative techniques for demonstrating the validity and applicability of the computational models. Computational models provide deterministic results based on deterministic and well-defined input, or stochastic results based on inputs defined by probability distributions. However, computational models cannot account for the effects of personnel, procedures, and equipment, resulting in ambiguity about the efficacy of inspections based on guidance from computational models only. In addition ambiguity arises when model inputs, such as the representation of realistic cracks, cannot be defined deterministically, probabilistically, or by intervals. In this work, Pacific Northwest National Laboratory demonstrates the ability of computational models to represent field measurements under known variabilities, and quantify the differences using maximum amplitude and power spectrum density metrics. Sensitivity studies are also conducted to quantify the effects of different input parameters on the simulation results.

Dib, Gerges↗
DOE OSTI2011

MAGNETARS VERSUS HIGH MAGNETIC FIELD PULSARS: A THEORETICAL INTERPRETATION OF THE APPARENT DICHOTOMY

Highly magnetized neutron stars (NSs) are characterized by a bewildering range of astrophysical manifestations. Here, building on our simulations of the evolution of magnetic stresses in the NS crust and its ensuing fractures, we explore in detail, for the middle-aged and old NSs, the dependence of starquake frequency and energetics on the relative strength of the poloidal (B{sub p}) and toroidal (B{sub tor}) components. We find that, for B{sub p} {approx}> 10{sup 14} G, since a strong crustal toroidal field B{sub tor} {approx} B{sub p} is quickly formed on a Hall timescale, the initial toroidal field needs to be B{sub tor} >> B{sub p} to have a clear influence on the outbursting behavior. For initial fields B{sub p} {approx}< 10{sup 14} G, it is very unlikely that a middle-aged (t {approx} 10{sup 5} years) NS shows any bursting activity. This study allows us to solve the apparent puzzle of how NSs with similar dipolar magnetic fields can behave in a remarkably different way: an outbursting 'magnetar' with a high X-ray luminosity, or a quiet, low-luminosity, 'high-B' radio pulsar. As an example, we consider the specific cases of the magnetar 1E 2259+586 and the radio pulsar PSR J1814-1744, which at present have a similar dipolar field {approx}6 Multiplication-Sign 10{sup 13} G. We determine for each object an initial magnetic field configuration that reproduces the observed timing parameters at their current age. The same two configurations also account for the differences in quiescent X-ray luminosity and for the 'magnetar/outbursting' behavior of 1E 2259+586 but not of PSR J1814-1744. We further use the theoretically predicted surface temperature distribution to compute the light curve for these objects. In the case of 1E 2259+586, for which data are available, our predicted temperature distribution gives rise to a pulse profile whose double-peaked nature and modulation level are consistent with the observations.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Multistage distributionally robust mixed-integer programming with decision-dependent moment-based ambiguity sets

We study multistage distributionally robust mixed-integer programs under endogenous uncertainty, where the probability distribution of stage-wise uncertainty depends on the decisions made in previous stages. We first consider two ambiguity sets defined by decision-dependent bounds on the first and second moments of uncertain parameters and by mean and covariance matrix that exactly match decision-dependent empirical ones, respectively. For both sets, we show that the subproblem in each stage can be recast as a mixed-integer linear program (MILP). Moreover, we extend the general moment-based ambiguity set in to the multistage decision-dependent setting, and derive mixed-integer semidefinite programming (MISDP) reformulations of stage-wise subproblems. We develop methods for attaining lower and upper bounds of the optimal objective value of the multistage MISDPs, and approximate them using a series of MILPs. We deploy the Stochastic Dual Dynamic integer Programming (SDDiP) method for solving the problem under the three ambiguity sets with risk-neutral or risk-averse objective functions, and conduct numerical studies on multistage facility-location instances having diverse sizes under different parameter and uncertainty settings. Furthermore, our results show that the SDDiP quickly finds optimal solutions for moderate-sized instances under the first two ambiguity sets, and also finds good approximate bounds for the multistage MISDPs derived under the third ambiguity set. We also demonstrate the efficacy of incorporating decision-dependent distributional ambiguity in multistage decision-making processes.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Challenges in reducing the computational time of QSTS simulations for distribution system analysis.

The rapid increase in penetration of distributed energy resources on the electric power distribution system has created a need for more comprehensive interconnection modelling and impact analysis. Unlike conventional scenario - based studies , quasi - static time - series (QSTS) simulation s can realistically model time - dependent voltage controllers and the diversity of potential impacts that can occur at different times of year . However, to accurately model a distribution system with all its controllable devices, a yearlong simulation at 1 - second resolution is often required , which could take conventional computers a computational time of 10 to 120 hours when an actual unbalanced distribution feeder is modeled . This computational burden is a clear l imitation to the adoption of QSTS simulation s in interconnection studies and for determining optimal control solutions for utility operations . Our ongoing research to improve the speed of QSTS simulation has revealed many unique aspects of distribution system modelling and sequential power flow analysis that make fast QSTS a very difficult problem to solve. In this report , the most relevant challenges in reducing the computational time of QSTS simulations are presented: number of power flows to solve, circuit complexity, time dependence between time steps, multiple valid power flow solutions, controllable element interactions, and extensive accurate simulation analysis.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2015

SU-E-T-574: Novel Chance-Constrained Optimization in Intensity-Modulated Proton Therapy Planning to Account for Range and Patient Setup Uncertainties

Purpose: We propose to apply a probabilistic framework, namely chanceconstrained optimization, in the intensity-modulated proton therapy (IMPT) planning subject to range and patient setup uncertainties. The purpose is to hedge against the influence of uncertainties and improve robustness of treatment plans. Methods: IMPT plans were generated for a typical prostate patient. Nine dose distributions are computed — the nominal one and one each for ±5mm setup uncertainties along three cardinal axes and for ±3.5% range uncertainty. These nine dose distributions are supplied to the solver CPLEX as chance constraints to explicitly control plan robustness under these representative uncertainty scenarios with certain probability. This probability is determined by the tolerance level. We make the chance-constrained model tractable by converting it to a mixed integer optimization problem. The quality of plans derived from this method is evaluated using dose-volume histogram (DVH) indices such as tumor dose homogeneity (D5% – D95%) and coverage (D95%) and normal tissue sparing like V70 of rectum, V65, and V40 of bladder. We also compare the results from this novel method with the conventional PTV-based method to further demonstrate its effectiveness Results: Our model can yield clinically acceptable plans within 50 seconds. The chance-constrained optimization produces IMPT plans with comparable target coverage, better target dose homogeneity, and better normal tissue sparing compared to the PTV-based optimization [D95% CTV: 67.9 vs 68.7 (Gy), D5% – D95% CTV: 11.9 vs 18 (Gy), V70 rectum: 0.0 % vs 0.33%, V65 bladder: 2.17% vs 9.33%, V40 bladder: 8.83% vs 21.83%]. It also simultaneously makes the plan more robust [Width of DVH band at D50%: 2.0 vs 10.0 (Gy)]. The tolerance level may be varied to control the tradeoff between plan robustness and quality. Conclusion: The chance-constrained optimization generates superior IMPT plan compared to the PTV-based optimization with explicit control of plan robustness. NIH/NCI K25CA168984, Eagles Cancer Research Career Development, The Lawrence W. and Marilyn W. Matteson Fund for Cancer Research, Mayo ASU Seed Grant, and The Kemper Marley Foundation.

62 RADIOLOGY AND NUCLEAR MEDICINE↗
DOE OSTI2018

Power-traffic coordinated operation for bi-peak shaving and bi-ramp smoothing – A hierarchical data-driven approach

The power distribution system and urban transportation system are two networked system bare their own operation constraints, such peak load in power systems and traffic congestion in transportation system. With the increasing number of electrical vehicles and charging/discharging stations, two systems are become tightly coupled. However, to optimize the two systems target using electrical vehicles as decision control variables cannot be easily solved using a uniformed optimization frame work. Thus we propose a hierarchical optimization approach to address this problem, which consists of a higher and a lower level. In the higher level, the power distribution system and urban transportation system are treated together to minimize the social cost. Meanwhile, the electrical vehicles and the charging/discharging stations are treated as customers to minimize their own expenditures. Then, an equilibrium is designed to determine the optimal charging/discharging price. In the lower level, the models of power distribution system and urban transportation system are developed to provide a detailed analysis. Specifically, in power distribution system, the three-phase unbalanced optimal power flow problem is relaxed with the semidefinite relaxation programming, and solved with alternating direction method of multiplier. A dynamic user equilibrium problem is formulated for the urban transportation system. For electrical vehicles, the state of charge is considered to optimize the charging/discharging schedule and reduce the impacts of power distribution systems. We conducted the simulation and numerical analysis using the IEEE 8500-bus distribution system and the Sioux Falls system with about 10,000 cars. The results demonstrate the feasibility and effectiveness of the proposed approach.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2012

Uncertainty Quantification of Hypothesis Testing for the Integrated Knowledge Engine

The Integrated Knowledge Engine (IKE) is a tool of Bayesian analysis, based on Bayesian Belief Networks or Bayesian networks for short. A Bayesian network is a graphical model (directed acyclic graph) that allows representing the probabilistic structure of many variables assuming a localized type of dependency called the Markov property. The Markov property in this instance makes any node or random variable to be independent of any non-descendant node given information about its parent. A direct consequence of this property is that it is relatively easy to incorporate new evidence and derive the appropriate consequences, which in general is not an easy or feasible task. Typically we use Bayesian networks as predictive models for a small subset of the variables, either the leave nodes or the root nodes. In IKE, since most applications deal with diagnostics, we are interested in predicting the likelihood of the root nodes given new observations on any of the children nodes. The root nodes represent the various possible outcomes of the analysis, and an important problem is to determine when we have gathered enough evidence to lean toward one of these particular outcomes. This document presents criteria to decide when the evidence gathered is sufficient to draw a particular conclusion or decide in favor of a particular outcome by quantifying the uncertainty in the conclusions that are drawn from the data. The material in this document is organized as follows: Section 2 presents briefly a forensics Bayesian network, and we explore evaluating the information provided by new evidence by looking first at the posterior distribution of the nodes of interest, and then at the corresponding posterior odds ratios. Section 3 presents a third alternative: Bayes Factors. In section 4 we finalize by showing the relation between the posterior odds ratios and Bayes factors and showing examples these cases, and in section 5 we conclude by providing clear guidelines of how to use these for the type of Bayesian networks used in IKE.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

TH-CD-209-04: Fuzzy Robust Optimization in Intensity-Modulated Proton Therapy Planning to Account for Range and Patient Setup Uncertainties

Purpose: We propose to apply a robust optimization model based on fuzzy-logic constraints in the intensity-modulated proton therapy (IMPT) planning subject to range and patient setup uncertainties. The purpose is to ensure the plan robustness under uncertainty and obtain the best trade-off between tumor dose coverage and organ-at-risk(OAR) sparing. Methods: Two IMPT plans were generated for 3 head-and-neck cancer patients: one used the planning target volume(PTV) method; the other used the fuzzy robust optimization method. In the latter method, nine dose distributions were computed - the nominal one and one each for ±3mm setup uncertainties along three cardinal axes and for ±3.5% range uncertainty. For tumors, these nine dose distributions were explicitly controlled by adding hard constraints with adjustable parameters. For OARs, fuzzy constraints that allow the dose to vary within a certain range were used so that the tumor dose distribution was guaranteed by minimum compromise of that of OARs. We rendered this model tractable by converting the fuzzy constraints to linear constraints. The plan quality was evaluated using dose-volume histogram(DVH) indices such as tumor dose coverage(D95%), homogeneity(D5%-D95%), plan robustness(DVH band at D95%), and OAR sparing like D1% of brain and D1% of brainstem. Results: Our model could yield clinically acceptable plans. The fuzzy-logic robust optimization method produced IMPT plans with comparable target dose coverage and homogeneity compared to the PTV method(unit: Gy[RBE]; average[min, max])(CTV D95%: 59 [52.7, 63.5] vs 53.5[46.4, 60.1], CTV D5% - D95%: 11.1[5.3, 18.6] vs 14.4[9.2, 21.5]). It also generated more robust plans(CTV DVH band at D95%: 3.8[1.2, 5.6] vs 11.5[6.2, 16.7]). The parameters of tumor constraints could be adjusted to control the tradeoff between tumor coverage and OAR sparing. Conclusion: The fuzzy-logic robust optimization generates superior IMPT with minimum compromise of OAR sparing. This research was supported by the National Cancer Institute Career Developmental Award K25CA168984, by the Fraternal Order of Eagles Cancer Research Fund Career Development Award, by The Lawrence W. and Marilyn W. Matteson Fund for Cancer Research, by Mayo Arizona State University Seed Grant, and by The Kemper Marley Foundation. eRA Person ID(s) for the Principal Investigator: 11017970 (Research Supported by National Institutes of Health)

60 APPLIED LIFE SCIENCES↗
DOE OSTI2019

DataFed

There is a need to easily and uniformly capture, access, enrich, organize, search, and collaborate on working scientific data sets across geographically distributed experiment, computation, and data analytics facilities. DataFed provides a set of services and applications enabling simple and uniform cross-facility scientific data management with federated (distributed) data repositories. This solution greatly reduces effort required by scientist to organize and/or move data between facilities/institutes, Reduces likelihood of misidentification and subsequent misuse of data, and improves scientific reproducibility through deep, domain-specific context and provenance capture.

Stansberry, DaleV.↗
DOE OSTI2018

Multivariate Analogs of Classical Univariate Discrete Distributions and Their Properties

Some discrete distributions such as Bernoulli, binomial, geometric, negative binomial, Poisson, Polya–Aeppli, and others play an important role in applied problems of probability theory and mathematical statistics. We propose a variant of a multivariate distribution whose components have a given univariate discrete distribution. In fact we consider some very general variant of the so-called reduction method. We find the explicit form of the mass function and generating function of such distribution and study their properties. We prove that our construction is unique in natural exponential families of distributions. Our results are the generalization and unification of many results of other authors.

97 MATHEMATICS AND COMPUTING↗