Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel, scaling”

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 703 records · Page 39
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

A performance spectrum for parallel computational frameworks that solve PDEs

Summary Important computational physics problems are often large‐scale in nature, and it is highly desirable to have robust and high performing computational frameworks that can quickly address these problems. However, it is no trivial task to determine whether a computational framework is performing efficiently or is scalable. The aim of this paper is to present various strategies for better understanding the performance of any parallel computational frameworks for solving PDEs. Important performance issues that negatively impact time‐to‐solution are discussed, and we propose a performance spectrum analysis that can enhance one's understanding of critical aforementioned performance issues. As proof of concept, we examine commonly used finite element simulation packages and software and apply the performance spectrum to quickly analyze the performance and scalability across various hardware platforms, software implementations, and numerical discretizations. It is shown that the proposed performance spectrum is a versatile performance model that is not only extendable to more complex PDEs such as hydrostatic ice sheet flow equations but also useful for understanding hardware performance in a massively parallel computing environment. Potential applications and future extensions of this work are also discussed.

Chang, J.↗
NASA NTRS2000

Testing New Programming Paradigms with NAS Parallel Benchmarks

Over the past decade, high performance computing has evolved rapidly, not only in hardware architectures but also with increasing complexity of real applications. Technologies have been developing to aim at scaling up to thousands of processors on both distributed and shared memory systems. Development of parallel programs on these computers is always a challenging task. Today, writing parallel programs with message passing (e.g. MPI) is the most popular way of achieving scalability and high performance. However, writing message passing programs is difficult and error prone. Recent years new effort has been made in defining new parallel programming paradigms. The best examples are: HPF (based on data parallelism) and OpenMP (based on shared memory parallelism). Both provide simple and clear extensions to sequential programs, thus greatly simplify the tedious tasks encountered in writing message passing programs. HPF is independent of memory hierarchy, however, due to the immaturity of compiler technology its performance is still questionable. Although use of parallel compiler directives is not new, OpenMP offers a portable solution in the shared-memory domain. Another important development involves the tremendous progress in the internet and its associated technology. Although still in its infancy, Java promisses portability in a heterogeneous environment and offers possibility to "compile once and run anywhere." In light of testing these new technologies, we implemented new parallel versions of the NAS Parallel Benchmarks (NPBs) with HPF and OpenMP directives, and extended the work with Java and Java-threads. The purpose of this study is to examine the effectiveness of alternative programming paradigms. NPBs consist of five kernels and three simulated applications that mimic the computation and data movement of large scale computational fluid dynamics (CFD) applications. We started with the serial version included in NPB2.3. Optimization of memory and cache usage was applied to several benchmarks, noticeably BT and SP, resulting in better sequential performance. In order to overcome the lack of an HPF performance model and guide the development of the HPF codes, we employed an empirical performance model for several primitives found in the benchmarks. We encountered a few limitations of HPF, such as lack of supporting the "REDISTRIBUTION" directive and no easy way to handle irregular computation. The parallelization with OpenMP directives was done at the outer-most loop level to achieve the largest granularity. The performance of six HPF and OpenMP benchmarks is compared with their MPI counterparts for the Class-A problem size in the figure in next page. These results were obtained on an SGI Origin2000 (195MHz) with MIPSpro-f77 compiler 7.2.1 for OpenMP and MPI codes and PGI pghpf-2.4.3 compiler with MPI interface for HPF programs.

Jin, H.↗
DOE OSTI2024

T RI M E ++: Multi-threaded triangular meshing in two dimensions

We present T RI M E ++, a multi-threaded software library designed for generating two-dimensional meshes for intricate geometric shapes using the Delaunay triangulation. Multi-threaded parallel computing is implemented throughout the meshing procedure, making it suitable for fast generation of large-scale meshes. Three iterative meshing algorithms are implemented: the DistMesh algorithm, the centroidal Voronoi diagram meshing, and a hybrid of the two. We compare the performance of the three meshing methods in T RI M E ++, and show that the hybrid method retains the advantages of the other two. The software library achieves significant parallel speedup when generating large-scale meshes containing between 10 4 to 10 7 points. T RI M E ++ can handle complicated geometries and generates adaptive meshes of high quality.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

A scalable Euler–Lagrange approach for multiphase flow simulation on spectral elements

Multiphase flow can be difficult to simulate with high accuracy due to the wide range of scales associated with various multiphase phenomena. These scales may range from the size of individual particles to the entire domain of interest. Traditionally, large scale systems can only be simulated using averaging approaches that filter out the locations of individual particles. In this work, the Euler–Lagrange method is used to simulate large-scale dense particle systems in which each individual particle is tracked. In order to accomplish this, the highly scalable spectral element code nek5000 has been extended to handle the multiple levels of multiphase coupling in these systems. These levels include what has been called one-, two-, and four-way coupling. Here, each level has been separated to detail the computational impact of each stage. A binned ghost particle algorithm has also been developed to efficiently handle the challenges of two- and four-way coupling in a parallel processing context. The algorithms and their implementations are then shown to scale to 65,536 Message Passing Interface (MPI) ranks in both the strong and weak limits. After this, validation is performed through simulation of a small-scale fluidized bed. Lastly, a large-scale fluidized bed is simulated with 65,536 MPI ranks and is able to capture the unique physics of the onset of fluidization.

Computer Science↗
DOE OSTI2019

Structural deformation and transformation of θ'-Al 2 Cu precipitate in Al matrix via interfacial diffusion

Metastable θ'-Al 2 Cu precipitates enhance mechanical properties of Al alloy, but gradually transform to θ phase due to diffusion related coarsening at higher temperatures (>200°C). To improve understanding of the phase instability and transformation of θ'-Al 2 Cu, we investigated the interfacial atomic mobility of θ'-Al2Cu precipitate in Al matrix via molecular dynamics (MD). To characterize the interfacial atomic mobility, diffusion properties (activation energy, diffusivity, and jump attempt frequency) with respect to atomic species, interfacial structure, and temperature are calculated using atomic trajectories from MD. For the enhanced accuracy of this analysis, especially at low temperatures, temporal scale of molecular dynamics is significantly extended by employing the parallel replica dynamics. This study, as the first MD investigation on atomic diffusion of θ'-Al 2 Cu in Al matrix, suggests and discusses (i) structural deformation activated at the semi-coherent θ'-Al 2 Cu/Al interfaces, (ii) transformation initiated at the edge of the coherent θ'-Al 2 Cu/Al interface, and (iii) directional dependence of diffusion properties. Overall, these findings are expected to be employed in a larger scale phase-field modeling of precipitate growth, and to contribute to the understanding of phase transformation process and the development of alloys with θ'-phase retained to higher temperatures.

36 MATERIALS SCIENCE↗
DOE OSTI2017

Multigrid methods with space–time concurrency

Here, we consider the comparison of multigrid methods for parabolic partial differential equations that allow space–time concurrency. With current trends in computer architectures leading towards systems with more, but not faster, processors, space–time concurrency is crucial for speeding up time-integration simulations. In contrast, traditional time-integration techniques impose serious limitations on parallel performance due to the sequential nature of the time-stepping approach, allowing spatial concurrency only. This paper considers the three basic options of multigrid algorithms on space–time grids that allow parallelism in space and time: coarsening in space and time, semicoarsening in the spatial dimensions, and semicoarsening in the temporal dimension. We develop parallel software and performance models to study the three methods at scales of up to 16K cores and introduce an extension of one of them for handling multistep time integration. We then discuss advantages and disadvantages of the different approaches and their benefit compared to traditional space-parallel algorithms with sequential time stepping on modern architectures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

A case study on parallel HDF5 dataset concatenation for high energy physics data analysis

In High Energy Physics (HEP), experimentalists generate large volumes of data that, when analyzed, helps us better understand the fundamental particles and their interactions. This data is often captured in many files of small size, creating a data management challenge for scientists. In order to better facilitate data management, transfer, and analysis on large scale platforms, it is advantageous to aggregate data further into a smaller number of larger files. However, this translation process can consume significant time and resources, and if performed incorrectly the resulting aggregated files can be inefficient for highly parallel access during analysis on large scale platforms. In this paper, we present our case study on parallel I/O strategies and HDF5 features for reducing data aggregation time, making effective use of compression, and ensuring efficient access to the resulting data during analysis at scale. We focus on NOvA detector data in this case study, a large-scale HEP experiment generating many terabytes of data. Here, the lessons learned from our case study inform the handling of similar datasets, thus expanding community knowledge related to this common data management task.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

ORBIT: Oak Ridge Base Foundation Model for Earth System Predictability

Earth system predictability is challenged by the complexity of environmental dynamics and the multitude of variables involved. Current AI foundation models, although advanced by leveraging large and heterogeneous data, are often constrained by their size and data integration, limiting their effectiveness in addressing the full range of Earth system prediction challenges. To overcome these limitations, we introduce the Oak Ridge Base Foundation Model for Earth System Predictability (ORBIT), an advanced vision transformer model that scales up to 113 billion parameters using a novel hybrid tensor-data orthogonal parallelism technique. As the largest model of its kind, ORBIT surpasses the current climate AI foundation model size by a thousandfold. Performance scaling tests conducted on the Frontier supercomputer have demonstrated that ORBIT achieves 684 petaFLOPS to 1.6 exaFLOPS sustained throughput, with scaling efficiency maintained at 41% to 85% across 49,152 AMD GPUs. These breakthroughs establish new advances in AI-driven climate modeling and demonstrate promise to significantly improve the Earth system predictability.

Wang, Xiao↗
DOE OSTI2026

Scale setting of SU⁡(𝑁) Yang–Mills theory, topology and large-𝑁 volume independence

We set the scale of SU⁡(𝑁) Yang-Mills theories for 𝑁 =3, 5, 8 and in the large-𝑁 limit via gradient flow, as a first step towards the computation of the large-𝑁 Λ-parameter using step scaling. We adopt twisted boundary conditions to achieve large-𝑁 volume reduction and the Parallel Tempering on Boundary Conditions algorithm to tame topological freezing. This setup allows accurate determinations of the gradient-flow scales down to lattice spacings as fine as ∼0.025 fm for all the explored values of 𝑁, a regime that has never been reached with ergodic algorithms. Moreover, we are able to precisely estimate the finite-size systematics related to topological freezing, and to show the suppression of finite-volume effects expected by virtue of large-𝑁 twisted volume reduction.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
NASA NTRS2014

Structural Health Monitoring on Turbine Engines Using Microwave Blade Tip Clearance Sensors

The ability to monitor the structural health of the rotating components, especially in the hot sections of turbine engines, is of major interest to aero community in improving engine safety and reliability. The use of instrumentation for these applications remains very challenging. It requires sensors and techniques that are highly accurate, are able to operate in a high temperature environment, and can detect minute changes and hidden flaws before catastrophic events occur. The National Aeronautics and Space Administration (NASA) has taken a lead role in the investigation of new sensor technologies and techniques for the in situ structural health monitoring of gas turbine engines. As part of this effort, microwave sensor technology has been investigated as a means of making high temperature non-contact blade tip clearance, blade tip timing, and blade vibration measurements for use in gas turbine engines. This paper presents a summary of key results and findings obtained from the evaluation of two different types of microwave sensors that have been investigated for use possible in structural health monitoring applications. The first is a microwave blade tip clearance sensor that has been evaluated on a large scale Axial Vane Fan, a subscale Turbofan, and more recently on sub-scale turbine engine like disks. The second is a novel microwave based blade vibration sensor that was also used in parallel with the microwave blade tip clearance sensors on the experiments with the sub-scale turbine engine disks.

Gas Turbine Engines↗
NASA NTRS1987

A VLSI Processor Engine to Support Calculation of the Vector Wave Equation

The development of a VLSI processor engine tailored to support computation of the near-field vector wave equation (VWE) is presented. Calculation of the near field is essential for evaluation of the electromagnetic fields generated by a complex radiating structure. The processor engine presented is optimized such that it takes advantage of the extensive parallelism inherent in the computational algorithm. The utilization of very large scale integration (VLSI) circuit technology and a customized reduced instruction set computer (RISC) architecture support development of a system of these processor engines, each of which is capable of performing the VWE algorithm calculation. By taking maximum advantage of the parallelism present in the calculation and dividing the problem among an array of these processor engines, the calculation can be computed in a realistic time frame.

J W Degroat↗
DOE OSTI2022

3-regular three-XORSAT planted solutions benchmark of classical and quantum heuristic optimizers

With current semiconductor technology reaching its physical limits, special-purpose hardware has emerged as an option to tackle specific computing-intensive challenges. Optimization in the form of solving quadratic unconstrained binary optimization problems, or equivalently Ising spin glasses, has been the focus of several new dedicated hardware platforms. These platforms come in many different flavors, from highly-efficient hardware implementations on digital-logic of established algorithms to proposals of analog hardware implementing new algorithms. In this work, we use a mapping of a specific class of linear equations whose solutions can be found efficiently, to a hard constraint satisfaction problem (three-regular three-XORSAT, or an Ising spin glass) with a 'golf-course' shaped energy landscape, to benchmark several of these different approaches. We perform a scaling and prefactor analysis of the performance of Fujitsu's digital annealer unit (DAU), the D-Wave advantage quantum annealer, a virtual MemComputing machine, Toshiba's simulated bifurcation machine (SBM), the SATonGPU algorithm from Bernaschi et al, and our implementation of parallel tempering. We identify the SATonGPU and DAU as currently having the smallest scaling exponent for this benchmark, with SATonGPU having a small scaling advantage and in addition having by far the smallest prefactor thanks to its use of massive parallelism. Furthermore, our work provides an objective assessment and a snapshot of the promise and limitations of dedicated optimization hardware relative to a particular class of optimization problems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2017

Frog: Asynchronous Graph Processing on GPU with Hybrid Coloring Model

GPUs have been increasingly used to accelerate graph processing for complicated computational problems regarding graph theory. Many parallel graph algorithms adopt the asynchronous computing model to accelerate the iterative convergence. Unfortunately, the consistent asynchronous computing requires locking or atomic operations, leading to significant penalties/overheads when implemented on GPUs. As such, coloring algorithm is adopted to separate the vertices with potential updating conflicts, guaranteeing the consistency/correctness of the parallel processing. Common coloring algorithms, however, may suffer from low parallelism because of a large number of colors generally required for processing a large-scale graph with billions of vertices. We propose a light-weight asynchronous processing framework called Frog with a preprocessing/hybrid coloring model. The fundamental idea is based on Pareto principle (or 80-20 rule) about coloring algorithms as we observed through masses of realworld graph coloring cases. We find that a majority of vertices (about 80%) are colored with only a few colors, such that they can be read and updated in a very high degree of parallelism without violating the sequential consistency. Accordingly, our solution separates the processing of the vertices based on the distribution of colors. In this work, we mainly answer three questions: (1) how to partition the vertices in a sparse graph with maximized parallelism, (2) how to process large-scale graphs that cannot fit into GPU memory, and (3) how to reduce the overhead of data transfers on PCIe while processing each partition. Here, we conduct experiments on real-world data (Amazon, DBLP, YouTube, RoadNet-CA, WikiTalk and Twitter) to evaluate our approach and make comparisons with well-known non-preprocessed (such as Totem, Medusa, MapGraph and Gunrock) and preprocessed (Cusha) approaches, by testing four classical algorithms (BFS, PageRank, SSSP and CC). On all the tested applications and datasets, Frog is able to significantly outperform existing GPU-based graph processing systems except Gunrock and MapGraph. MapGraph gets better performance than Frog when running BFS on RoadNet-CA. The comparison between Gunrock and Frog is inconclusive. Frog can outperform Gunrock more than 1.04X when running PageRank and SSSP, while the advantage of Frog is not obvious when running BFS and CC on some datasets especially for RoadNet-CA.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Acceleration and sensitivity analysis of lattice kinetic Monte Carlo simulations using parallel processing and rate constant rescaling

Kinetic Monte Carlo (KMC) simulation provides insights into catalytic reactions unobtainable with either experiments or mean-field microkinetic models. Sensitivity analysis of KMC models assesses the robustness of the predictions to parametric perturbations and identifies rate determining steps in a chemical reaction network. Stiffness in the chemical reaction network, a ubiquitous feature, demands lengthy run times for KMC models and renders efficient sensitivity analysis based on the likelihood ratio method unusable. We address the challenge of efficiently conducting KMC simulations and performing accurate sensitivity analysis in systems with unknown time scales by employing two acceleration techniques: rate constant rescaling and parallel processing. We develop statistical criteria that ensure sufficient sampling of non-equilibrium steady state conditions. Our approach provides the twofold benefit of accelerating the simulation itself and enabling likelihood ratio sensitivity analysis, which provides further speedup relative to finite difference sensitivity analysis. As a result, the likelihood ratio method can be applied to real chemistry. We apply our methodology to the water-gas shift reaction on Pt(111).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

Massively Parallel Simulation of Large Molecular Systems with Long-Range Interactions Final Report CRADA No. TC-0297-92-B

The purpose of this CRADA project is to develop and implement a large-scale molecular dynamics model for floppy molecules with long-ranged interaction on massively parallel supercomputers and accelerate the use of parallel supercomputers within the biomolecular and polymer industries. Within three years of the project, the partners will: create a new molecular dynamics computer simulation code (LAMMPS) from existing codes at SAndia National Laboratory-Albuquerque (SNL-A) and Lawrence Livermore National Laboratory (LLNL) and new algorithms and code developed during the CRADA project; design the resulting code to optimally use the emerging generation of massively parallel supercomputers; and apply the code to the solution of biomolecular problem suggested by Bristol-Myers Squibb (BMS) and polymer problems suggested by DuPont.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Accelerating and Parallelizing Lagrangian Simulations of Mixing-Limited Reactive Transport

Recent advances in random walk particle tracking have enabled direct simulation of mixing and reactions by allowing the particles to interact with each other using a multipoint mass transfer scheme. The mass transfer scheme allows separation of mixing and spreading processes, among other advantages, but it is computationally expensive because its speed depends on the number of interacting particle pairs. This note explores methods for relieving the computational bottleneck caused by the mass transfer step, and we use these algorithms to develop a new parallel, interacting particle model. The new model is a combination of a sparse search algorithm and a novel domain decomposition scheme, both of which offer significant speedup relative to the reference case—even when they are executed serially. We combine the strengths of these methods to create a parallel particle scheme that is highly accurate and efficient with run times that scale as 1/P for a fixed number of particles, where P is the number of computational cores (equivalently, subdomains, in this work) being used. Here, the new parallel model is a significant advance because it enables efficient simulation of large particle ensembles that are needed for environmental simulations and also because it can naturally pair with parallel geochemical solvers to create a practical Lagrangian tool for simulating mixing and reactions in complex chemical systems.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2017

Onset of a Large Ejective Solar Eruption from a Typical Coronal-jet-base Field Configuration

Utilizing multiwavelength observations and magnetic field data from the Solar Dynamics Observatory ( SDO )/Atmospheric Imaging Assembly (AIA), SDO /Helioseismic and Magnetic Imager (HMI), the Geostationary Operational Environmental Satellite ( GOES ), and RHESSI , we investigate a large-scale ejective solar eruption of 2014 December 18 from active region NOAA 12241. This event produced a distinctive “three-ribbon” flare, having two parallel ribbons corresponding to the ribbons of a standard two-ribbon flare, and a larger-scale third quasi-circular ribbon offset from the other two. There are two components to this eruptive event. First, a flux rope forms above a strong-field polarity inversion line and erupts and grows as the parallel ribbons turn on, grow, and spread apart from that polarity inversion line; this evolution is consistent with the mechanism of tether-cutting reconnection for eruptions. Second, the eruption of the arcade that has the erupting flux rope in its core undergoes magnetic reconnection at the null point of a fan dome that envelops the erupting arcade, resulting in formation of the quasi-circular ribbon; this is consistent with the breakout reconnection mechanism for eruptions. We find that the parallel ribbons begin well before (∼12 minutes) the onset of the circular ribbon, indicating that tether-cutting reconnection (or a non-ideal MHD instability) initiated this event, rather than breakout reconnection. The overall setup for this large-scale eruption (diameter of the circular ribbon ∼10{sup 5} km) is analogous to that of coronal jets (base size ∼10{sup 4} km), many of which, according to recent findings, result from eruptions of small-scale “minifilaments.” Thus these findings confirm that eruptions of sheared-core magnetic arcades seated in fan–spine null-point magnetic topology happen on a wide range of size scales on the Sun.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS1994

Independent pixel and Monte Carlo estimates of stratocumulus albedo

Monte Carlo radiative transfer methods are employed here to estimate the plane-parallel albedo bias for marine stratocumulus clouds. This is the bias in estimates of the mesoscale-average albedo, which arises from the assumption that cloud liquid water is uniformly distributed. The authors compare such estimates with those based on a more realistic distribution generated from a fractal model of marine stratocumulus clouds belonging to the class of 'bounded cascade' models. In this model the cloud top and base are fixed, so that all variations in cloud shape are ignored. The model generates random variations in liquid water along a single horizontal direction, forming fractal cloud streets while conserving the total liquid water in the cloud field. The model reproduces the mean, variance, and skewness of the vertically integrated cloud liquid water, as well as its observed wavenumber spectrum, which is approximately a power law. The Monte Carlo method keeps track of the three-dimensional paths solar photons take through the cloud field, using a vectorized implementation of a direct technique. The simplifications in the cloud field studied here allow the computations to be accelerated. The Monte Carlo results are compared to those of the independent pixel approximation, which neglects net horizontal photon transport. Differences between the Monte Carlo and independent pixel estimates of the mesoscale-average albedo are on the order of 1% for conservative scattering, while the plane-parallel bias itself is an order of magnitude larger. As cloud absorption increases, the independent pixel approximation agrees even more closely with the Monte Carlo estimates. This result holds for a wide range of sun angles and aspect ratios. Thus, horizontal photon transport can be safely neglected in estimates of the area-average flux for such cloud models. This result relies on the rapid falloff of the wavenumber spectrum of stratocumulus, which ensures that the smaller-scale variability, where the radiative transfer is more three-dimensional, contributes less to the plane-parallel albedo bias than the larger scales, which are more variable. The lack of significant three-dimensional effects also relies on the assumption of a relatively simple geometry. Even with these assumptions, the independent pixel approximation is accurate only for fluxes averaged over large horizontal areas, many photon mean free paths in diameter, and not for local radiance values, which depend strongly on the interaction between neighboring cloud elements.

Cahalan, Robert F.↗