SEARCH · Engineering Papers

Results for “parallel machines”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 613 records · Page 34

All sources NASA NTRS DOE OSTI

DOE OSTI2020

Understanding performance variability in standard and pipelined parallel Krylov solvers

In this work, we collect data from runs of Krylov subspace methods and pipelined Krylov algorithms in an effort to understand and model the impact of machine noise and other sources of variability on performance. We find large variability of Krylov iterations between compute nodes for standard methods that is reduced in pipelined algorithms, directly supporting conjecture, as well as large variation between statistical distributions of runtimes across iterations. Based on these results, we improve upon a previously introduced nondeterministic performance model by allowing iterations to fluctuate over time. We present our data from runs of various Krylov algorithms across multiple platforms as well as our updated non-stationary model that provides good agreement with observations. We also suggest how it can be used as a predictive tool.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Parthenon—a performance portable block-structured adaptive mesh refinement framework

On the path to exascale the landscape of computer device architectures and corresponding programming models has become much more diverse. While various low-level performance portable programming models are available, support at the application level lacks behind. To address this issue, we present the performance portable block-structured adaptive mesh refinement (AMR) framework Parthenon, derived from the well-tested and widely used Athena++ astrophysical magnetohydrodynamics code, but generalized to serve as the foundation for a variety of downstream multi-physics codes. Parthenon adopts the Kokkos programming model, and provides various levels of abstractions from multidimensional variables, to packages defining and separating components, to launching of parallel compute kernels. Parthenon allocates all data in device memory to reduce data movement, supports the logical packing of variables and mesh blocks to reduce kernel launch overhead, and employs one-sided, asynchronous MPI calls to reduce communication overhead in multi-node simulations. Using a hydrodynamics miniapp, we demonstrate weak and strong scaling on various architectures including AMD and NVIDIA GPUs, Intel and AMD x86 CPUs, IBM Power9 CPUs, as well as Fujitsu A64FX CPUs. At the largest scale on Frontier (the first TOP500 exascale machine), the miniapp reaches a total of 1.7 × 10 13 zone-cycles/s on 9216 nodes (73,728 logical GPUs) at [Formula: see text] weak scaling parallel efficiency (starting from a single node). In combination with being an open, collaborative project, this makes Parthenon an ideal framework to target exascale simulations in which the downstream developers can focus on their specific application rather than on the complexity of handling massively-parallel, device-accelerated AMR.

97 MATHEMATICS AND COMPUTING↗

NASA NTRS2004

Parallel Performance Characterization of Columbia

Using a collection of benchmark problems of increasing levels of realism and computational effort, we will characterize the strengths and limitations of the 10,240 processor Columbia system to deliver supercomputing value to application scientists. Scientists need to be able to determine if and how they can utilize Columbia to carry extreme workloads, either in terms of ultra-large applications that cannot be run otherwise (capability), or in terms of very large ensembles of medium-scale applications to populate response matrices (capacity). We select existing application benchmarks that scale from a small number of processors to the entire machine, and that highlight different issues in running supercomputing-calss applicaions, such as the various types of memory access, file I/O, inter- and intra-node communications and parallelization paradigms. http://www.nas.nasa.gov/Software/NPB/

Biswas, Rupak↗

DOE OSTI2023

In-depth analysis on parallel processing patterns for high-performance Dataframes

The Data Science domain has expanded monumentally in both research and industry communities during the past decade, predominantly owing to the Big Data revolution. Artificial Intelligence (AI) and Machine Learning (ML) are bringing more complexities to data engineering applications, which are now integrated into data processing pipelines to process terabytes of data. Typically, a significant amount of time is spent on data preprocessing in these pipelines, and hence improving its efficiency directly impacts the overall pipeline performance. The community has recently embraced the concept of Dataframes as the de-facto data structure for data representation and manipulation. However, the most widely used serial Dataframes today (R, pandas) experience performance limitations while working on even moderately large data sets. We believe that there is plenty of room for improvement by taking a look at this problem from a high-performance computing point of view. In a prior publication, we presented a set of parallel processing patterns for distributed dataframe operators and the reference runtime implementation, Cylon. In this paper, we are expanding on the initial concept by introducing a cost model for evaluating the said patterns. Furthermore, we evaluate the performance of Cylon on the ORNL Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Machine learning-based optimization of air-cooled heat sinks

Machine learning-based models using Artificial Neural Network (ANN) and greedy search algorithm are used to optimize air-cooled parallel plate-finned heat sinks (PPFHSs) subjected to laminar flow over an extensive range of design parameters. Here, the thermal and hydraulic performances of PPFHSs are represented by heat transfer coefficient (h) and pressure drop (ΔP), respectively. Optimization objectives for PPFHS designs can vary from industry to industry depending on their design priorities. The present study proposes a novel and generalized optimization method that defines practical optimization objectives and provides an accurate optimization process to design effective PPFHSs for a wide range of industrial applications with different design requirements. Three optimization objectives are presented in this study: (i) the largest h PΔ, (ii) the largest h within a specified maximum allowed flow rate, and (iii) the lowest weight that maximizes h for operation within the maximum allowed flow rate. While the shortcoming of the first objective is demonstrated, the other two objectives are found to be suitable for designing effective heat sinks (HSs) across different applications. Results suggest a promising trend from the third objective to develop HSs with ~ 37-68% lower weight, 80-85% reduced ΔP, and negligible penalty in h compared with optimized HSs obtained from the second objective. However, since the third objective leads to HSs with thinner fins, structural analysis should be performed to ensure reliable operation of the HSs.

42 ENGINEERING↗

DOE OSTI2023

Exploring model complexity in machine learned potentials for simulated properties

Abstract Machine learning (ML) enables the development of interatomic potentials with the accuracy of first principles methods while retaining the speed and parallel efficiency of empirical potentials. While ML potentials traditionally use atom-centered descriptors as inputs, different models such as linear regression and neural networks map descriptors to atomic energies and forces. This begs the question: what is the improvement in accuracy due to model complexity irrespective of descriptors? We curate three datasets to investigate this question in terms of ab initio energy and force errors: (1) solid and liquid silicon, (2) gallium nitride, and (3) the superionic conductor Li $$_{10}$$ 10 Ge(PS $$_{6}$$ 6 ) $$_{2}$$ 2 (LGPS). We further investigate how these errors affect simulated properties and verify if the improvement in fitting errors corresponds to measurable improvement in property prediction. By assessing different models, we observe correlations between fitting quantity (e.g. atomic force) error and simulated property error with respect to ab initio values. Graphical abstract

Rohskopf, A. (ORCID:0000000227128296)↗

NASA NTRS1933

The Theory of the Strandgren Cyclogiro

This report provides a description of the Standgren flying machine which consists of two wheels with a certain number of equidistant blades arranged around the horizontal axis of rotation and parallel to this axis. Each blade is fixed so as to be able to feather about an axis parallel to its span at the same time as it turns about the general axis of rotation. A general theory of wheels with blades rotating about a transverse axis is presented.

Strandgren, C B↗

NASA NTRS1993

Parallel integer sorting with medium and fine-scale parallelism

Two new parallel integer sorting algorithms, queue-sort and barrel-sort, are presented and analyzed in detail. These algorithms do not have optimal parallel complexity, yet they show very good performance in practice. Queue-sort designed for fine-scale parallel architectures which allow the queueing of multiple messages to the same destination. Barrel-sort is designed for medium-scale parallel architectures with a high message passing overhead. The performance results from the implementation of queue-sort on a Connection Machine CM-2 and barrel-sort on a 128 processor iPSC/860 are given. The two implementations are found to be comparable in performance but not as good as a fully vectorized bucket sort on the Cray YMP.

Dagum, Leonardo↗

NASA NTRSDate not supplied

GPU Implementation of the OVERFLOW CFD Code

The high-performance computing (HPC) landscape is quickly changing to systems where most of the performance comes from specialized chips, specifically graphics processing units (GPUs). Such GPU systems are throughput machines, where efficient use of the GPU often requires code refactoring to expose a few orders of magnitude more fine grain parallelism than was previously used on the CPU. Recent modifications to OVERFLOW, an overset, structured grid, computational fluid dynamics flow solver, written in Fortran will be presented. These modifications include both code modernization efforts and algorithmic changes to enable OVERFLOW to efficiently utilize GPUs. Many of these algorithmic changes would likely also be applicable for other structured grid, stencil-based codes wanting to utilize GPUs. The capabilities that have been ported to run on the GPUs are presented, along with the performance gains of the GPU version relative the CPU version of OVERFLOW.

GPU Programming↗

NASA NTRSDate not supplied

GPU Implementation of the OVERFLOW CFD Code

GPU Programming↗

NASA NTRS1994

An implementation of a tree code on a SIMD, parallel computer

We describe a fast tree algorithm for gravitational N-body simulation on SIMD parallel computers. The tree construction uses fast, parallel sorts. The sorted lists are recursively divided along their x, y and z coordinates. This data structure is a completely balanced tree (i.e., each particle is paired with exactly one other particle) and maintains good spatial locality. An implementation of this tree-building algorithm on a 16k processor Maspar MP-1 performs well and constitutes only a small fraction (approximately 15%) of the entire cycle of finding the accelerations. Each node in the tree is treated as a monopole. The tree search and the summation of accelerations also perform well. During the tree search, node data that is needed from another processor is simply fetched. Roughly 55% of the tree search time is spent in communications between processors. We apply the code to two problems of astrophysical interest. The first is a simulation of the close passage of two gravitationally, interacting, disk galaxies using 65,636 particles. We also simulate the formation of structure in an expanding, model universe using 1,048,576 particles. Our code attains speeds comparable to one head of a Cray Y-MP, so single instruction, multiple data (SIMD) type computers can be used for these simulations. The cost/performance ratio for SIMD machines like the Maspar MP-1 make them an extremely attractive alternative to either vector processors or large multiple instruction, multiple data (MIMD) type parallel computers. With further optimizations (e.g., more careful load balancing), speeds in excess of today's vector processing computers should be possible.

Olson, Kevin M.↗

NASA NTRS1985

Evaluation of existing and proposed computer architectures for future ground-based systems

Parallel processing architectures and techniques used in current supercomputers are described and projections are made of future advances. Presently, the von Neumann sequential processing pattern has been accelerated by having separate I/O processors, interleaved memories, wide memories, independent functional units and pipelining. Recent supercomputers have featured single-input, multiple data stream architectures, which have different processors for performing various operations (vector or pipeline processors). Multiple input, multiple data stream machines have also been developed. Data flow techniques, wherein program instructions are activated only when data are available, are expected to play a large role in future supercomputers, along with increased parallel processor arrays. The enhanced operational speeds are essential for adequately treating data from future spacecraft remote sensing instruments such as the Thematic Mapper.

Schulbach, C.↗

DOE OSTI2023

Implementing machine learning methods on QICK hardware for qubit readout & control

Quantum readout and control is a fundamental aspect of quantum computing that requires accurate measurement of qubit states. Errors emerge in all stages, from initialization to readout, and identifying errors in post-processing necessitates resource-intensive statistical analysis. In our work, we use a lightweight fully-connected neural network (NN) to classify states of a transmon system with no prior processing. Our NN accelerator yields higher fidelities (92%) than the classical matched filter method (84%). By exploiting the natural parallelism of NNs and their placement near the source of data on field-programmable gate arrays (FPGAs), we can achieve ultra-low latency on the Quantum Instrumentation Control Kit (QICK). Integrating machine learning methods on QICK opens several pathways for efficient real-time processing of quantum circuits.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

NASA NTRS1998

Parallel 3D Multi-Stage Simulation of a Turbofan Engine

A 3D multistage simulation of each component of a modern GE Turbofan engine has been made. An axisymmetric view of this engine is presented in the document. This includes a fan, booster rig, high pressure compressor rig, high pressure turbine rig and a low pressure turbine rig. In the near future, all components will be run in a single calculation for a solution of 49 blade rows. The simulation exploits the use of parallel computations by using two levels of parallelism. Each blade row is run in parallel and each blade row grid is decomposed into several domains and run in parallel. 20 processors are used for the 4 blade row analysis. The average passage approach developed by John Adamczyk at NASA Lewis Research Center has been further developed and parallelized. This is APNASA Version A. It is a Navier-Stokes solver using a 4-stage explicit Runge-Kutta time marching scheme with variable time steps and residual smoothing for convergence acceleration. It has an implicit K-E turbulence model which uses an ADI solver to factor the matrix. Between 50 and 100 explicit time steps are solved before a blade row body force is calculated and exchanged with the other blade rows. This outer iteration has been coined a "flip." Efforts have been made to make the solver linearly scaleable with the number of blade rows. Enough flips are run (between 50 and 200) so the solution in the entire machine is not changing. The K-E equations are generally solved every other explicit time step. One of the key requirements in the development of the parallel code was to make the parallel solution exactly (bit for bit) match the serial solution. This has helped isolate many small parallel bugs and guarantee the parallelization was done correctly. The domain decomposition is done only in the axial direction since the number of points axially is much larger than the other two directions. This code uses MPI for message passing. The parallel speed up of the solver portion (no 1/0 or body force calculation) for a grid which has 227 points axially.

Turner, Mark G.↗

DOE OSTI2025

Spotlight: efficient automated global optimization in rietveld analysis of diffraction data

Performing reliable Rietveld analysis on tens or hundreds of powder diffraction datasets from parametric or time-resolved experiments often poses a bottleneck in extracting meaningful results from the data. While automated analysis of data has recently been demonstrated, high temperature annealing studies, during which phase transformations occur and lattice parameters may change due to repartitioning of elements, are prime examples where automation by a simple phase identification from a database of room temperature structures or automation by sequential refinements is likely to fail. To enable reliable, efficient, automated Rietveld analysis, we present a Python package named Spotlight , building on established Rietveld packages such as MAUD, GSAS , or GSAS-II , which extends the refinement of best fit parameters to a global optimization using an ensemble of optimizers leveraging hierarchical parallel execution on high-performance computing clusters. Spotlight further enables the efficient design of refinement plans through the iterative automated machine-learning of a surrogate for the refinement on which the global optimizations are performed until results from the surrogate converge to the response surface data. We demonstrate Spotlight with the analysis of uranium molybdenum and Ti–6Al–4V datasets, as well as in two open-source tutorials analyzing aluminium oxide and lead sulphate.

36 MATERIALS SCIENCE↗

DOE OSTI2024

PaRSEC: Scalability, flexibility, and hybrid architecture support for task-based applications in ECP

This paper highlights the most significant enhancements made to PaRSEC, a scalable task-based runtime system designed for hybrid machines, during the Exascale Computing Project (ECP). The enhancements focus on expanding the capabilities of PaRSEC to address the evolving landscape of parallel computing. Notable achievements include the integration of support for three major types of accelerators (NVIDIA, AMD, and Intel GPUs), the refinement and increased flexibility of the communication subsystem, and the introduction of new programming interfaces tailored for irregular applications. Additionally, the project resulted in the development of powerful debugging and performance analysis tools aimed at assisting users in understanding and optimizing their applications. We present a comprehensive demonstration of these advancements through a series of benchmarks and applications within ECP and beyond, thereby showcasing the enhanced capabilities of PaRSEC across the diverse architectures within the ECP, providing valuable insights into the runtime system’s adaptability and performance across varied computing environments.

Bouteiller, Aurelien↗

NASA NTRS1971

Ion-wave current instabilities and anomalous resistivity.

A theory of ion-wave current instabilities which takes into account, in a self-consistent manner, the inhomogeneities generated by field-aligned currents in a collisionless plasma is presented. Diamagnetic current associated with the current-produced density gradient is included in the distribution. The theory predicts that for a given frequency, the current threshold for ion-wave current instabilities is, in general, much below the threshold of ion acoustic instability in a uniform plasma as given by Fried and Gould. The current threshold is essentially zero in the limit when the ion Landau damping effect is negligible, or equivalently, the ion-wave current instabilities are absolute if there are no limitations on the wavelength. This is true even in the absence of externally applied density gradients in contrast to Kadomtsev's drift-wave results. For dimensions of interest in laboratory plasmas, the predicted linear growth rate increases with increasing longitudinal wavelength and with decreasing wavelength parallel to the diamagnetic current. Under the conditions of an experiment on anomalous resistivity (the dimensions but not the geometry of the machine have been considered), there is good agreement between the predicted onset of ion-wave current instabilities and the experimental data on the onset of anomalous resistivity.

Kan, J. R.↗

NASA NTRS1984

Modelling algorithm execution time on processor arrays

A model for the execution time of parallel algorithms on processor arrays is described. The model is validated for the conjugate gradient algorithm on the eight processor Finite Element Machine at NASA Langley Research Center. Model predictions are also included for this algorithm on a larger array as the number of processors and system parameters are varied.

Adams, L. M.↗