Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed parallelization”

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 631 records · Page 35

Mesh Algorithms for PDE with Sieve I: Mesh Distribution

We have developed a new programming framework, called Sieve, to support parallel numerical partial differential equation(s) (PDE) algorithms operating over distributed meshes. We have also developed a reference implementation of Sieve in C++ as a library of generic algorithms operating on distributed containers conforming to the Sieve interface. Sieve makes instances of the incidence relation, or arrows, the conceptual first-class objects represented in the containers. Further, generic algorithms acting on this arrow container are systematically used to provide natural geometric operations on the topology and also, through duality, on the data. Finally, coverings and duality are used to encode not only individual meshes, but all types of hierarchies underlying PDE data structures, including multigrid and mesh partitions. In order to demonstrate the usefulness of the framework, we show how the mesh partition data can be represented and manipulated using the same fundamental mechanisms used to represent meshes. We present the complete description of an algorithm to encode a mesh partition and then distribute a mesh, which is independent of the mesh dimension, element shape, or embedding. Moreover, data associated with the mesh can be similarly distributed with exactly the same algorithm. The use of a high level of abstraction within the Sieve leads to several benefits in terms of code reuse, simplicity, and extensibility. We discuss these benefits and compare our approach to other existing mesh libraries.

Knepley, Matthew G.↗

SU-C-207A-06: On-Line Beam Range Verification with Multiple Scanning Particle Beams: Initial Feasibility Study with Simulations

Purpose: To investigate the feasibility and requirement for intra-fraction on-line multiple scanning particle beam range verifications (BRVs) with in-situ PET imaging, which is beyond the current single-beam BRV with extra factors that will affect the BR measurement accuracy, such as beam diameter, separation between beams, and different image counts at different BRV positions. Methods: We simulated a 110-MeV proton beam with 5-mm diameter irradiating a uniform PMMA phantom by GATE simulation, which generated nuclear interaction-induced positrons. In this preliminary study, we simply duplicated these positrons and placed them next to the initial protons to approximately mimic the two spatially separated positron distributions produced by two beams parallel to each other but with different beam ranges. These positrons were then imaged by a PET (∼2-mm resolution, 10% sensitivity, 320×320×128 mm^3 FOV) with different acquisition times. We calculated the positron activity ranges (ARs) from reconstructed PET images and compared them with the corresponding ARs of original positron distributions. Results: Without further image data processing and correction, the preliminary study show the errors between the measured and original ARs varied from 0.2 mm to 2.3 mm as center-to-center separations and range differences were in the range of 8–12 mm and 2–8 mm respectively, indicating the accuracy of AR measurement strongly depends on the beam separations and range differences. In addition, it is feasible to achieve ≤ 1.0-mm accuracy for both beams with 1-min PET acquisition and 12 mm beam separation. Conclusion: This study shows that the overlap between the positron distributions from multiple scanning beams can significantly impact the accuracy of BRVs of distributed particle beams and need to be further addressed beyond the established method of single-beam BRV, but it also indicates the feasibility to achieve accurate on-line multi-beam BRV with further improved method.

60 APPLIED LIFE SCIENCES↗

Parallel Variable Population Multi-Objective Optimizer (pvpmoo) v1.0

This is a parallel variable population multi-objective optimizer with an adaptive unified differential evolution algorithm or a genetic algorithm. It can also be used for single objective optimization. Some features of this code include: 1) The population size varies from generation to generation to save the total # of objective function evaluations. 2) The population is uniformly distributed to a number of parallel processors for simultaneous objective function evaluation. 3) The objective function evaluation can be attained from an external simulation program with control variables in its input file and objectives calculated from its output files. 4) The optimizer includes an adaptive unified differential evolution algorithm and a real value genetic algorithm. The parameters in the unified differential evolution algorithm can be chosen to attain any mutation schemes in the published literature.

Qiang, Ji↗

REPRODUCTION OF THE OBSERVED TWO-COMPONENT MAGNETIC HELICITY IN SOLAR WIND TURBULENCE BY A SUPERPOSITION OF PARALLEL AND OBLIQUE ALFVEN WAVES

The angular distribution of the normalized reduced magnetic helicity density ({sigma} r{sub m}) in solar wind turbulence reveals two components of distinct polarity in different angle ranges. This kind of two-component {sigma}{sup r}{sub m} may indicate the possible wave modes and power spectral densities (PSDs) of the turbulent fluctuations. Here we model the measured angular distribution of {sigma}{sup r}{sub m} by assuming a PSD distribution for Alfven fluctuations in wavevector space, and then fit the model results to the observations by adjusting the pattern of the PSD distribution. It is found that the two-component form of the PSD, which has a major and minor component close to k and k{sub ||}, respectively, seems to be responsible for the observed two-component {sigma}{sup r}{sub m}. On the other hand, both an isotropic PSD and a PSD with only a single component bending toward k fail to reproduce the observations. Moreover, it is shown that the effect of gradual balance between outward and inward wave-energy fluxes with decreasing spatial scale needs to be considered in order to reproduce the observed diminishing of |{sigma}{sup r}{sub m}| at shorter scales. Therefore, we suggest that the observed two-component {sigma}{sup r}{sub m} in the solar wind turbulence may be due to a superposition of Alfven waves with quasi-perpendicular (major part) and quasi-parallel (minor part) propagation. The waves seem to become gradually balanced toward shorter scales.

79 ASTRONOMY AND ASTROPHYSICS↗

Expressing Parallelism with ROOT

The need for processing the ever-increasing amount of data generated by the LHC experiments in a more efficient way has motivated ROOT to further develop its support for parallelism. Such support is being tackled both for shared-memory and distributed-memory environments. The incarnations of the aforementioned parallelism are multi-threading, multi-processing and cluster-wide executions. In the area of multi-threading, we discuss the new implicit parallelism and related interfaces, as well as the new building blocks to safely operate with ROOT objects in a multi-threaded environment. Regarding multi-processing, we review the new MultiProc framework, comparing it with similar tools (e.g. multiprocessing module in Python). Finally, as an alternative to PROOF for cluster-wide executions, we introduce the efforts on integrating ROOT with state-of-the-art distributed data processing technologies like Spark, both in terms of programming model and runtime design (with EOS as one of the main components). For all the levels of parallelism, we discuss, based on real-life examples and measurements, how our proposals can increase the productivity of scientists.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

A Full-Stack Exploration of Language-Based Parallelism in Fortran 2023

This poster explores native parallel features in Fortran 2023 through the lens of supporting applications with libraries, compilers, and parallel runtimes. The language revision informally named Fortran 2008 introduced parallelism in the form of Single Program Multiple Data (SPMD) execution with two broad feature sets: (1) loop-level parallelism via do concurrent and (2) a Partitioned Global Address Space (PGAS) comprised of distributed “coarray” data structures. Fortran’s native parallelism has demonstrated high performance [1] and reduced the burden of inserting what sometimes amounts to more directives than code. Several compilers support both feature sets, typically by translating do concurrent into serial do loops annotated by parallel directives and by translating SPMD/PGAS features into direct calls to a communication library. Our research focuses primarily on two questions: (1) can the compiler’s parallel runtime library be developed in the language being compiled (Fortran) and (2) can we define an interface to the runtime that liberates compilers from being hardwired to one runtime and vice versa. We are answering these questions by developing the Parallel Runtime Interface for Fortran (PRIF) [2] and the Co-Array Fortran Framework of Efficient Interfaces to Network Environments (Caffeine) [3]. Caffeine is initially targeting adoption by LLVM Flang, a new open-source Fortran compiler developed by a broad community in industry, academia, and government labs. We are also exploring the use of these features in Inference-Engine, a deep learning library designed to facilitate neural network training and inference for high-performance computing applications written in modern Fortran.

Rasmussen, Katherine↗

Using the Global Arrays Toolkit to Reimplement NumPy for Distributed Computation

Global Arrays (GA) is a software system from Pacific Northwest National Laboratory that enables an efficient, portable, and parallel shared memory programming interface to manipulate distributed defense arrays. Using a combination of GA and NumPy, we have reimplemented NumPy as a distributed drop-in replacement called Global Arrays in NumPy (GAiN). Scalability studies will be presented showing the utility of developing serial NumPy codes which can later run on more capable clusters or supercomputers.

Global Arrays↗

Transition from the twinning induced plasticity to the γ-ε transformation induced plasticity in a high manganese steel

Neutron-diffraction investigation on the deformation of a Fe-18Mn-3Si-0.6C-0.4Al steel reveals that the twinning is mainly dominant at a stress below 900 MPa, above which the twinning behavior is largely outweighed by phase transformation from the face-centered-cubic (FCC) γ-austenite to the hexagonal-close-packed (HCP) ε-martensite. In the deformed grains, the distribution of ε-martensite is parallel with the twin boundary. Both the well-known {111} γ //{0001} ε relationship and an additional {111} γ //{11–21} ε orientation relation were identified. After the phase transformation, both phases deformed further with the lattice rotating around an axis perpendicular to the tensile direction. Transition from the twinning dominant behavior to the phase transformation induced plasticity effect is explained by the grain orientation dependence of the effective stacking fault energy (ESFE). An asymmetric and inverse relationship between the width of stacking faults (SF) and the ESFE is obtained from the density functional theory (DFT). The tensile stress always increases proportionally with the SF width in the twinning favorable grains and thus a decreasing ESFE. The replacement of twinning activities by phase transformation could be due to both the nucleation site of ε-martensite at the SF of the twin boundary and the decrease of the ESFE in grains which originally favors the twinning.

Materials Science↗

Development of Steady-State and Dynamic Mass and Energy Constrained Neural Networks for Distributed Chemical Systems Using Noisy Transient Data

The paper presents the development of algorithms for mass and energy constrained neural network models that can exactly conserve the overall mass and energy of distributed chemical process systems, even though the noisy transient data used for optimal model training violate the same. In contrast to approximately satisfying mass and energy balance constraints of a system by soft penalization of objective function, algorithms have been developed for solving equality-constrained nonlinear optimization problems, thus providing the guarantee of exactly satisfying the system mass and energy conservation laws. For developing dynamic mass-energy constrained network models for distributed systems, hybrid series and parallel dynamic-static neural networks have been leveraged. The developed algorithms for solving both the training and forward problems are validated using both steady-state and dynamic data in the presence of various noise characteristics. The developed data-driven algorithms are flexible to exactly satisfy mass and energy balance constraints for dynamic chemical processes if the system holdup information is available. The proposed network structures and algorithms are applied to the development of data-driven lumped and distributed models of an adiabatic superheater/reheater system, a nonisothermal continuous stirred tank reactor, as well as an electrically heated plug-flow reactor system where one form of energy gets transformed to another. It has been observed that the mass-energy constrained neural networks yield a root mean squared error of <1% with respect to the system truth for the case studies evaluated in this work.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Domain Decomposition of a Constructive Solid Geometry Monte Carlo Transport Code

Domain decomposition has been implemented in a Constructive Solid Geometry (CSG) Monte Carlo neutron transport code. Previous methods to parallelize a CSG code relied entirely on particle parallelism; but in our approach we distribute the geometry as well as the particles across processors. This enables calculations whose geometric description is larger than what could fit in memory of a single processor, thus it must be distributed across processors. In addition to enabling very large calculations, we show that domain decomposition can speed up calculations compared to particle parallelism alone. We also show results of a calculation of the proposed Laser Inertial-Confinement Fusion-Fission Energy (LIFE) facility, which has 5.6 million CSG parts.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

GRID-BASED EXPLORATION OF COSMOLOGICAL PARAMETER SPACE WITH SNAKE

We present a fully parallelized grid-based parameter estimation algorithm for investigating multidimensional likelihoods called Snake, and apply it to cosmological parameter estimation. The basic idea is to map out the likelihood grid-cell by grid-cell according to decreasing likelihood, and stop when a certain threshold has been reached. This approach improves vastly on the 'curse of dimensionality' problem plaguing standard grid-based parameter estimation simply by disregarding grid cells with negligible likelihood. The main advantages of this method compared to standard Metropolis-Hastings Markov Chain Monte Carlo methods include (1) trivial extraction of arbitrary conditional distributions; (2) direct access to Bayesian evidences; (3) better sampling of the tails of the distribution; and (4) nearly perfect parallelization scaling. The main disadvantage is, as in the case of brute-force grid-based evaluation, a dependency on the number of parameters, N{sub par}. One of the main goals of the present paper is to determine how large N{sub par} can be, while still maintaining reasonable computational efficiency; we find that N{sub par} = 12 is well within the capabilities of the method. The performance of the code is tested by comparing cosmological parameters estimated using Snake and the WMAP-7 data with those obtained using CosmoMC, the current standard code in the field. We find fully consistent results, with similar computational expenses, but shorter wall time due to the perfect parallelization scheme.

79 ASTRONOMY AND ASTROPHYSICS↗

Final Report: Migration Mechanisms for Large-scale Parallel Applications

Process migration is the ability to transfer a process from one machine to another. It is a useful facility in distributed computing environments, especially as computing devices become more pervasive and Internet access becomes more ubiquitous. The potential benefits of process migration, among others, are fault resilience by migrating processes off of faulty hosts, data access locality by migrating processes closer to the data, better system response time by migrating processes closer to users, dynamic load balancing by migrating processes to less loaded hosts, and improved service availability and administration by migrating processes before host maintenance so that applications can continue to run with minimal downtime. Although process migration provides substantial potential benefits and many approaches have been considered, achieving transparent process migration functionality has been difficult in practice. To address this problem, our work has designed, implemented, and evaluated new and powerful transparent process checkpoint-restart and migration mechanisms for desktop, server, and parallel applications that operate across heterogeneous cluster and mobile computing environments. A key aspect of this work has been to introduce lightweight operating system virtualization to provide processes with private, virtual namespaces that decouple and isolate processes from dependencies on the host operating system instance. This decoupling enables processes to be transparently checkpointed and migrated without modifying, recompiling, or relinking applications or the operating system. Building on this lightweight operating system virtualization approach, we have developed novel technologies that enable (1) coordinated, consistent checkpoint-restart and migration of multiple processes, (2) fast checkpointing of process and file system state to enable restart of multiple parallel execution environments and time travel, (3) process migration across heterogeneous software environments, (4) network checkpoint-restart and migration of distributed and parallel applications, (5) a utility computing infrastructure for mobile desktop cloud computing based on process checkpoint-restart and migration functionality, (6) a process migration security architecture for protecting applications and infrastructure from denial-of-service attacks, and (7) a checkpoint-restart mobile computing system using portable storage devices.

99 GENERAL AND MISCELLANEOUS↗

The residual zonal flows in anisotropic tokamak plasmas

The gyro-kinetic equation is analytically solved based on the anisotropic two-temperature distribution, in which the ions' parallel temperature is a flux function while the perpendicular temperature depends on the poloidal angle. The residual level of collisionless zonal flows (ZFs) is derived and calculated in the large aspect circular limit. Our result shows that the anisotropy plays a remarkable role in determining the residual value of ZFs. Even weak anisotropy can significantly change the residual level.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Scalable Distributed Memory Community Detection Using Vite

Graph clustering, popularly known as community detection, is a fundamental graph operation used in several applications of relevance for network analysis and cybersecurity. Community detection is an operation of partitioning a network into an arbitrary number of “communities” such that each com munity represents a tightly-knit group of nodes with relatively sparser connections to the rest of the nodes in the network. The need to compute clustering on large-scale networks necessitates the development of efficient parallel algorithms capable of exploiting modern parallel architectures. However, due to their irregular and inherently sequential nature, many of the current algorithms for community detection are challenging to parallelize. In response to the 2018 Graph Challenge, we present Vite which is a distributed-memory parallel community detection method. Vite is a parallelization of the algorithmic template provided by a widely-used serial method (Louvain). Alongside, it uses several heuristics aimed at significantly improving parallelism and performance, while preserving output quality. Using the inputs from the 2018 Graph Challenge (static and streaming), we demonstrate superior performance and high quality solutions based on four parallelization heuristics.

community detection↗

Doppler coherence imaging and tomography of flows in tokamak plasmas (invited)

This article describes the results of spatial heterodyne Doppler ''coherence imaging'' of carbon ion flows in the divertor region of the DIII-D tokamak. Spatially encoded interferometric projections of doubly ionized carbon emission at 465 nm have been demodulated and tomographically inverted to obtain the spatial distribution of the carbon ion parallel flow and emissivity. The operating principles of the new instruments are described, and the link between measured properties and line integrals of the flow field are established. An iterative simultaneous arithmetic reconstruction procedure is applied to invert the interferometric phase shift projections, and the reconstructed parallel flow field amplitudes are found to be in reasonable agreement with UEDGE modeling.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Optimal design of 3D borehole seismic arrays for microearthquake monitoring in anisotropic media during stimulations in the EGS collab project

Multiple U.S. national laboratories, universities and industrial collaborators are conducting collaborative research under the EGS Collab project supported by the U.S. Department of Energy, to understand the fracture creation and imaging during fracturing in enhanced geothermal systems. Microearthquake hypocenter locations and focal mechanisms are used to monitor hydraulic fracturing growth and characterization at the EGS Collab experimental site at the Sanford Underground Research Facility using seismic receivers in multiple monitoring wells. In this study, we develop a methodology for optimal design a 3D borehole seismic array for cost-effective seismic monitoring in anisotropic media using not only the relationship between receiver distributions and standard deviation errors of microearthquake hypocenter locations, but also that between receiver distributions and focal-mechanism inversion errors. Our results indicate that microearthquake hypocenter locations and focal mechanisms can be reasonably well reconstructed for the EGS Collab Experiment I using six monitoring wells, including four fracture-parallel monitoring wells and two orthogonal wells. Lastly, eight seismic receivers evenly distributed in four parallel monitoring wells or twelve receivers in all six monitoring wells are required for hypocenter location, and twelve receivers evenly distributed in six wells or sixteen receivers in four wells are needed for focal-mechanism inversion.

58 GEOSCIENCES↗

Hierarchical power distribution in large scale computing systems

Techniques for managing power distribution amongst processors in a massively parallel computer architecture are disclosed. The techniques utilize a hierarchy that organizes the various processors of the massively parallel computer architecture. The hierarchy groups numbers of the processors at the lowest level. When processors complete tasks, the power assigned to those processors is distributed to other processors in the same group so that the performance of those processors can be increased. Hierarchical organization simplifies the calculations required for determining how and when to distribute power, because when tasks are complete and power is available for distribution, a relatively small number of processors are available for consideration to receive that power. The number of processors that are grouped together can be adjusted in real time based on performance factors to improve the trade-off between calculation speed and power distribution efficacy.

42 ENGINEERING↗

Chemical and Structural Alterations in the Amorphous Structure of Obsidian due to Nanolites

Obsidian is volcanic glass that results from the rapid cooling of silica-rich melt. Nanoscale crystallites precipitate out of the melt prior to solidification and remain embedded in the amorphous matrix. These crystallites provide information on the flow kinetics and composition of the melt. Due to the sparsity and size of nanolites, studies often focus on supramicron crystallites. This research takes advantage of the conchoidal fracture of obsidian by knapping samples with nanometer-thin edges for transmission electron microscopy characterization. Nanolites in the amorphous matrix are studied using energy-dispersive spectroscopy (EDS) and electron diffraction. Certain alkali and alkaline-earth cations exhibit patterns of depletion near Fe-oxide nanolites. EDS is used to identify nanolites and variations in the composition of the matrix. Parallel beam diffraction and radial distribution function analysis of nearest-neighbor distances determine average bond lengths in the matrix near nanolites, showing that nanolites influence the nearby short-range ordering and atomic character of the matrix. Analysis reveals decreased mean nearest-neighbor distances in the matrix adjacent to nanolites compared to the bulk. Our methods exhibit the required sensitivity to detect variations in the composition and structure near nanolites, and our findings indicate that obsidian nanolites contribute to quantifiable localized changes in the amorphous structure.

36 MATERIALS SCIENCE↗