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 1,495 records · Page 83
All sourcesNASA NTRSDOE OSTI
NASA NTRS1993

MEDUSA - An overset grid flow solver for network-based parallel computer systems

Continuing improvement in processing speed has made it feasible to solve the Reynolds-Averaged Navier-Stokes equations for simple three-dimensional flows on advanced workstations. Combining multiple workstations into a network-based heterogeneous parallel computer allows the application of programming principles learned on MIMD (Multiple Instruction Multiple Data) distributed memory parallel computers to the solution of larger problems. An overset-grid flow solution code has been developed which uses a cluster of workstations as a network-based parallel computer. Inter-process communication is provided by the Parallel Virtual Machine (PVM) software. Solution speed equivalent to one-third of a Cray-YMP processor has been achieved from a cluster of nine commonly used engineering workstation processors. Load imbalance and communication overhead are the principal impediments to parallel efficiency in this application.

Smith, Merritt H.↗
NASA NTRS1972

Optimal shield mass distribution for space radiation protection

Computational methods have been developed and successfully used for determining the optimum distribution of space radiation shielding on geometrically complex space vehicles. These methods have been incorporated in computer program SWORD for dose evaluation in complex geometry, and iteratively calculating the optimum distribution for (minimum) shield mass satisfying multiple acute and protected dose constraints associated with each of several body organs.

Billings, M. P.↗
DOE OSTI2015

Future computing platforms for science in a power constrained era

Power consumption will be a key constraint on the future growth of Distributed High Throughput Computing (DHTC) as used by High Energy Physics (HEP). This makes performance-per-watt a crucial metric for selecting cost-efficient computing solutions. For this paper, we have done a wide survey of current and emerging architectures becoming available on the market including x86-64 variants, ARMv7 32-bit, ARMv8 64-bit, Many-Core and GPU solutions, as well as newer System-on-Chip (SoC) solutions. We compare performance and energy efficiency using an evolving set of standardized HEP-related benchmarks and power measurement techniques we have been developing. In conclusion, we evaluate the potential for use of such computing solutions in the context of DHTC systems, such as the Worldwide LHC Computing Grid (WLCG).

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

SMALE: Enhancing Scalability of Machine Learning Algorithms on Extreme-Scale Computing Platforms

Deployment and execution of machine learning tasks on extreme-scale computing platforms face several significant technical challenges: 1) High computing cost incurred by dense networks – The computing workload of deep networks with densely-connected topology increases rapidly with the network size, imposing a non-scalable computing model of extreme-scale computing platforms; 2) Non-optimized workload distribution – Many advanced deep learning algorithms, e.g., sparsification and irregular net-work topology, produce very unbalanced workload distribution on extreme-scale computing platforms. The computation efficiency is greatly hindered by the incurred data and computation redundancies as well as long tails of the node with extensive workload; 3) Constraints in data movement and I/O bottle-neck – Inter-node data movement in extreme-scale computing platforms are associated with high energy and latency costs, and subject to the constraints of I/O bandwidth; and 4) Generalization of algorithm realization and acceleration on computing platforms – The large varieties of machine learning algorithms and structures of extreme-scale computing platforms make the derivation of a generalized algorithm realization and acceleration method very challenging, which, however, is the requirement by domain scientists and interested users. We call the above challenges Smale’s Problems in Machine Learning and Understanding for High-Performance Computing Scientific Discovery. The objective of our three-year research project is to develop a holistic innovation set at structure, assembly, and acceleration layers of machine learning algorithms to address the above challenges in algorithm deployment and execution. Three tasks are particularly performed, including: At the algorithm structure level, we investigate the techniques that can structurally sparsify on the topology of deep networks for computing workload reduction. We also study clustering and pruning techniques that can optimize the workload distributions over the extreme-scale computing platforms; At the algorithm assembly level, we derive a unified learning framework for unsupervised transfer learning and dynamic growing capabilities. Novel training methods are also exploited to enhance the training efficiency of the proposed framework; At the algorithm acceleration level, we will develop a series of techniques that can accelerate the computation of sparse matrix operations, which are one of the core executions in deep learning and optimize memory access of the concerned platforms. Our proposed techniques attack the fundamental problems in machine learning algorithms running on extreme-scale computing platforms by vertically integrating the solutions at three closely entangled layers, paving the long-term scaling path of machine learning applications under DOE context. Three tasks corresponding to the above respective research orientations are performed during the three-year project period with our collaborators at ORNL. The outcome of the proposed project is anticipated to form a holistic solution set of novel algorithms and network topologies, efficient training techniques, and fast acceleration methods to promote the computing scalability of the machine learning applications of particular interest to DOE.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1983

Equivalent absorption coefficients generated from frequency probability distributions

A flexible and computationally accurate method of treating aerosol scattering in spectral regions in which gaseous absorption is important is described. In the method, line-by-line absorption coefficients are computed as a function of pressure, temperature, and absorber gas for the spectral region of interest. The coefficients are sorted into a probability distribution which is converted into a cumulative probability distribution, which in turn can be inverted due to its monotonic nature. The inverted distribution is a smooth curve giving the absorption coefficient as a function of an independent variable on the domain. The frequency integration of the radiative transfer equation can then be performed by a quadrature technique with values of the absorption coefficient determined from the inverted distribution curve. The method is illustrated by applying it to the 9.6 micron band of ozone.

Ackerman, T. P.↗
DOE OSTI2018

Proton Isovector Helicity Distribution on the Lattice at Physical Pion Mass

We present a state-of-the-art calculation of the isovector quark helicity Bjorken-x distribution in the proton using lattice-QCD ensembles at the physical pion mass. We compute quasi-distributions at proton momenta P z ϵ {2.2; 2.6; 3.0} GeV on the lattice, and match them systematically to the physical parton distribution using large-momentum effective theory (LaMET).We reach an unprecedented precision through high statistics in simulations, large-momentum proton matrix elements, and control of excited-state contamination. The resulting distribution with combined statistical and systematic errors is in agreement with the latest phenomenological analysis of the spin-dependent experimental data; in particular, Δ$\bar{u}$(x) > Δ$\bar{d}$(x).

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2018

Scalable deep text comprehension for Cancer surveillance on high-performance computing

Background: Deep Learning (DL) has advanced the state-of-the-art capabilities in bioinformatics applications which has resulted in trends of increasingly sophisticated and computationally demanding models trained by larger and larger data sets. This vastly increased computational demand challenges the feasibility of conducting cutting-edge research. One solution is to distribute the vast computational workload across multiple computing cluster nodes with data parallelism algorithms. In this study, we used a High-Performance Computing environment and implemented the Downpour Stochastic Gradient Descent algorithm for data parallelism to train a Convolutional Neural Network (CNN) for the natural language processing task of information extraction from a massive dataset of cancer pathology reports. We evaluated the scalability improvements using data parallelism training and the Titan supercomputer at Oak Ridge Leadership Computing Facility. To evaluate scalability, we used different numbers of worker nodes and performed a set of experiments comparing the effects of different training batch sizes and optimizer functions.Results: We found that Adadelta would consistently converge at a lower validation loss, though requiring over twice as many training epochs as the fastest converging optimizer, RMSProp. The Adam optimizer consistently achieved a close 2nd place minimum validation loss significantly faster; using a batch size of 16 and 32 allowed the network to converge in only 4.5 training epochs.Conclusions: We demonstrated that the networked training process is scalable across multiple compute nodes communicating with message passing interface while achieving higher classification accuracy compared to a traditional machine learning algorithm.

Qiu, John X.↗
NASA NTRS1981

Energetic electrons and plasma waves associated with a solar type III radio burst

Detailed in situ observations from the ISEE 3 spacecraft of energetic electrons, plasma waves, and radio emission for the type II solar radio burst of February 17, 1979, are presented. The reduced, one-dimensional electron distribution function is constructed as a function of time. Since the faster electrons arrive before the slower ones, a bump on tail distribution forms which is unstable to the growth of Langmuir waves. The plasma wave growth computed from the distribution function agrees well with the observed onset of the Langmuir waves, and there is qualitative agreement between variations in the plasma wave levels and in the development of regions of positive slope in the function. The evolution of the function, however, predicts far higher plasma wave levels than those observed. The maximum levels observed are approximately equal to the threshold for nonlinear wave processes, such as oscillation two-stream instability and soliton collapse.

Lin, R. P.↗
NASA NTRS1987

Computing architecture for telerobots in earth orbit

Based on generic operational and computational requirements associated with the control of telerobots in earth orbit, a multibus-based distributed but integrated computing architecture is proposed. An experimental system of that kind under development at the Jet Propulsion Laboratory (JPL) is briefly described. It uses Intel Multibus I at both control station and remote robot (telerobot) computing nodes. An essential element within each multibus is a Unified (or Universal) Computer Control Subsystem (UCCS) for telerobot and control station motor components. The two multibus-based computing nodes can be linked by parallel or high speed serial links for real-time data transmission and for closing the real-time bilateral (force-reflecting) control loop between telerobot and control station. The experimental system is briefly commented, followed by a brief discussion of future development plans and possibilities.

Bejczy, A. K.↗
NASA NTRS1989

Numerical methods in Markov chain modeling

Several methods for computing stationary probability distributions of Markov chains are described and compared. The main linear algebra problem consists of computing an eigenvector of a sparse, usually nonsymmetric, matrix associated with a known eigenvalue. It can also be cast as a problem of solving a homogeneous singular linear system. Several methods based on combinations of Krylov subspace techniques are presented. The performance of these methods on some realistic problems are compared.

Philippe, Bernard↗
DOE OSTI2019

A Case Study of MPI Over Long Distance Connections

Scientific workflows are increasingly being distributed across wide-area networks, and their code executions are expected to span across geographically dispersed computing systems. MPI has been extensively used to support communications for distributed computations, typically, over compute clusters and high-performance systems within a single facility. We present a case study of performance of MPI basic operations over long distance connections, wherein TCP is used for the underlying transport. We present measurements of execution times of MPI codes that utilize MPI Sendrecv operations over emulated 10Gbps connections with 0-366ms round-trip times, including the longest one spanning the globe. They demonstrate that basic MPI codes can be sustained over long distance connections under external packet loss rates up to 10%. They also highlight the qualitative effects of losses which manifest as increased execution times as a consequence of TCP’s loss recovery process.

Rao, Nageswara S.↗
NASA NTRS2001

Automation of Data Traffic Control on DSM Architecture

The design of distributed shared memory (DSM) computers liberates users from the duty to distribute data across processors and allows for the incremental development of parallel programs using, for example, OpenMP or Java threads. DSM architecture greatly simplifies the development of parallel programs having good performance on a few processors. However, to achieve a good program scalability on DSM computers requires that the user understand data flow in the application and use various techniques to avoid data traffic congestions. In this paper we discuss a number of such techniques, including data blocking, data placement, data transposition and page size control and evaluate their efficiency on the NAS (NASA Advanced Supercomputing) Parallel Benchmarks. We also present a tool which automates the detection of constructs causing data congestions in Fortran array oriented codes and advises the user on code transformations for improving data traffic in the application.

Frumkin, Michael↗
DOE OSTI2020

Scalable Knowledge Graph Analytics at 136 Petaflop/s

We are motivated by newly proposed methods for data mining large-scale corpora of scholarly publications, such as the full biomedical literature, which may consist of tens of millions of papers spanning decades of research. In this setting, analysts seek to discover how concepts relate to one another. They construct graph representations from annotated text databases and then formulate the relationship-mining problem as one of computing all-pairs shortest paths (APSP), which becomes a significant bottleneck. In this context, we present a new high-performance algorithm and implementation of the Floyd-Warshall algorithm for distributed-memory parallel computers accelerated by GPUs, which we call DSNAPSHOT (Distributed Accelerated Semiring All-Pairs Shortest Path). For our largest experiments, we ran DSNAPSHOT on a connected input graph with millions of vertices using 4, 096nodes (24,576GPUs) of the Oak Ridge National Laboratory's Summit supercomputer system. We find DSNAPSHOT achieves a sustained performance of 136×1015 floating-point operations per second (136petaflop/s) at a parallel efficiency of 90% under weak scaling and, in absolute speed, 70% of the best possible performance given our computation (in the single-precision tropical semiring or “min-plus” algebra). Looking forward, we believe this novel capability will enable the mining of scholarly knowledge corpora when embedded and integrated into artificial intelligence-driven natural language processing workflows at scale.

Kannan, Ramakrishnan {ramki}↗
NASA NTRS1999

The Effects of Buoyancy and Dilution on the Structure and Lift-off of Coflow Laminar Diffusion Flames

The ability to predict the coupled effects of complex transport phenomena with detailed chemical kinetics in diffusion flames is critical in the modeling of turbulent reacting flows and in understanding the processes by which soot formation and radiative transfer take place. In addition, an understanding of the factors that affect flame extinction in diffusion flames is critical in the suppression of fires and in improving engine efficiency. The goal of our characterizations of coflow laminar diffusion flames is to bring to microgravity the multidimensional diagnostic tools available in normal gravity, and in so doing provide a broader understanding of the successes and limitations of current combustion models. This will lead to a more detailed understanding of the interaction of convection, diffusion and chemistry in both buoyant and nonbuoyant environments. As a sensitive marker of changes in the flame shape, the number densities of excited-state CH (A(exp 2)delta, denoted CH*), and excited-state OH (A(exp 2)Sigma, denoted OH*) are measured in mu-g and normal gravity. Two-dimensional CH* and OH* number densities are deconvoluted from line-of-sight chemiluminescence measurements made on the NASA KC-135 reduced-gravity aircraft. Measured signal levels are calibrated, post-flight, with Rayleigh scattering. Although CH* and OH* kinetics are not well understood, the CH*, OH*, and ground-state CH distributions are spatially coincident in the flame anchoring region. Therefore, the ground-state CH distribution, which is easily computed, and the readily measured CH*/OH* distributions can be used to provide a consistent and convenient way of measuring lift-off height and flame shape in the diffusion flame under investigation. Given that the fuel composition affects flame chemistry and that buoyancy influences the velocity profile of the flow, we have the opportunity to computationally and experimentally study the roles of fluids and chemistry. In performing this microgravity study, improvements to the computational model have been made and new calculations performed for a range of gravity and flow conditions. Furthermore, modifications to the experimental approach were required as a consequence of the constraints imposed by existing microgravity facilities. Results from the computations and experiments are presented in the following sections.

Walsh, Kevin T.↗
DOE OSTI2018

Resilient Cyber Infrastructure for the Minimum Wind Curtailment Remedial Control Scheme

Integration of Distributed Energy Resources (DER) has been significantly increasing in the electric power grid. One of the predominant renewable energy sources, wind energy, possess a significant challenge for grid operations, due to its intermittency, uncertainty and variability. In changing operating conditions with wind uncertainty, part of wind may need to be curtailed for the safe utilization of the transmission system. In order to minimize the wind power curtailment and to keep the power flow on a transmission line under limits, a cyber resilient Distributed Remedial Action Scheme (DRAS) is developed in this paper. The RAS is formulated as an optimization problem, considering the real power flow constraints of the transmission lines and given wind variability. In this paper, DRAS is developed to operate even with failures in the computing nodes to make it cyber-resilient. Distributed simplex algorithm is used for optimization in the distributed RAS. The DRAS is implemented in a decentralized platform called Resilient Information Architecture Platform for Smart Grid (RIAPS) to enable cyber resiliency of DRAS. The effectiveness of the proposed approach is validated through offline simulations of the New England 39 bus system and online simulations of IEEE 14 bus test system. In conclusion, a cyber physical test bed utilizing Real Time Digital Simulator (RTDS), Phasor Measurement Units (PMUs), and BeagleBones has been used for the online simulation and validation of the developed cyber-resilient RAS.

17 WIND ENERGY↗
NASA NTRS1992

Distributed and parallel Ada and the Ada 9X recommendations

Recently, the DoD has sponsored work towards a new version of Ada, intended to support the construction of distributed systems. The revised version, often called Ada 9X, will become the new standard sometimes in the 1990s. It is intended that Ada 9X should provide language features giving limited support for distributed system construction. The requirements for such features are given. Many of the most advanced computer applications involve embedded systems that are comprised of parallel processors or networks of distributed computers. If Ada is to become the widely adopted language envisioned by many, it is essential that suitable compilers and tools be available to facilitate the creation of distributed and parallel Ada programs for these applications. The major languages issues impacting distributed and parallel programming are reviewed, and some principles upon which distributed/parallel language systems should be built are suggested. Based upon these, alternative language concepts for distributed/parallel programming are analyzed.

Volz, Richard A.↗
DOE OSTI2017

Imaging of the Li spatial distribution within V 2 O 5 cathode in a coin cell by neutron computed tomography

An understanding of Lithium (Li) spatial distribution within the electrodes of a Li-ion cell, during charge and discharge cycles, is essential to optimize the electrode parameters for increased performance under cycling. In this work, it is demonstrated that the spatial distribution of Li within Vanadium Pentoxide (V 2 O 5 ) electrodes of a small coin cell can be imaged by neutron computed tomography. The neutron attenuation data has been used to construct the three-dimensional Li spatial images. Specifically, it is shown that there is sufficient neutron imaging contrast between lithiated and delithiated regions of V 2 O 5 electrode making it possible to map Li distributions even in small electrodes with thicknesses <1 mm. The images reveal that the Li spatial distribution is inhomogeneous and a relatively higher C-rate leads to more non-uniform Li distribution after Li insertion. The non-uniform distribution suggests the limitation of Li diffusion within the electrode during the lithiation process under the relatively high cycling rates.

25 ENERGY STORAGE↗