Search NASA⌕ Search

SEARCH · Search NASA

Results for “data parallel”

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 901 records · Page 50
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Array-based Hierarchical Mesh Generation in Parallel

In this paper, we describe an array-based hierarchical mesh generation capability through uniform refinement of unstructured meshes for efficient solution of PDE's using finite element methods and multigrid solvers. A multi-degree, multi-dimensional and multi-level framework is designed to generate the nested hierarchies from an initial mesh that can be used for a number of purposes such as multi-level methods to generating large meshes. The capability is developed under the parallel mesh framework “Mesh Oriented dAtaBase” a.k.a MOAB. We describe the underlying data structures and algorithms to generate such hierarchies and present numerical results for computational efficiency and mesh quality. In conclusion, we also present results to demonstrate the applicability of the developed capability to a multigrid finite-element solver.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Rethinking key–value store for parallel I/O optimization

Key-value stores are being widely used as the storage system for large-scale internet services and cloud storage systems. However, they are rarely used in HPC systems, where parallel file systems are the dominant storage solution. In this study, we examine the architecture differences and performance characteristics of parallel file systems and key-value stores. We propose using key-value stores to optimize overall Input/Output (I/O) performance, especially for workloads that parallel file systems cannot handle well, such as the cases with intense data synchronization or heavy metadata operations. We conducted experiments with several synthetic benchmarks, an I/O benchmark, and a real application. We modeled the performance of these two systems using collected data from our experiments, and we provide a predictive method to identify which system offers better I/O performance given a specific workload. The results show that we can optimize the I/O performance in HPC systems by utilizing key-value stores.

Hyperdex↗
DOE OSTI2015

Optimizing Irregular Applications for Energy and Performance on the Tilera Many-core Architecture

Optimizing applications simultaneously for energy and performance is a complex problem. High performance, parallel, irregular applications are notoriously hard to optimize due to their data-dependent memory accesses, lack of structured locality and complex data structures and code patterns. Irregular kernels are growing in importance in applications such as machine learning, graph analytics and combinatorial scientific computing. Performance- and energy-efficient implementation of these kernels on modern, energy efficient, multicore and many-core platforms is therefore an important and challenging problem. We present results from optimizing two irregular applications { the Louvain method for community detection (Grappolo), and high-performance conjugate gradient (HPCCG) { on the Tilera many-core system. We have significantly extended MIT's OpenTuner auto-tuning framework to conduct a detailed study of platform-independent and platform-specific optimizations to improve performance as well as reduce total energy consumption. We explore the optimization design space along three dimensions: memory layout schemes, compiler-based code transformations, and optimization of parallel loop schedules. Using auto-tuning, we demonstrate whole node energy savings of up to 41% relative to a baseline instantiation, and up to 31% relative to manually optimized variants.

irregular applications, energy opttimization, many↗
DOE OSTI2023

Online and Scalable Data Compression Pipeline with Guarantees on Quantities of Interest

Data compression is becoming critical for data-intensive scientific applications. Scientists require compression techniques that accurately preserve derived quantities of interest (QoIs). Prior work has shown that a pipeline can be built to guarantee error on the primary data (PD) within user-defined bounds and achieve near-floating point QoI errors. In this paper, we present novel computational approaches for accelerating the pipeline and demonstrate results that enable concurrent execution of compression in parallel with the simulation nodes. This allows compression, including the writing of the required compression data, for the previous time step to be completed while the simulation proceeds with the current time step. Overall, the approach presented in this paper results in a 6–8 times improvement in computational overhead compared to previous work. These results were obtained using data generated by a large-scale fusion code called XGC, which produces hundreds of terabytes of data in a single day.

Banerjee, Tania↗
DOE OSTI2022

Shocks in the stacked Sunyaev-Zel’dovich profiles of clusters II: Measurements from SPT-SZ + Planck Compton- y map

ABSTRACT We search for the signature of cosmological shocks in stacked gas pressure profiles of galaxy clusters using data from the South Pole Telescope (SPT). Specifically, we stack the latest Compton-y maps from the 2500 deg2 SPT-SZ survey on the locations of clusters identified in that same data set. The sample contains 516 clusters with mean mass $\langle M_{\rm 200m}\rangle = 10^{14.9} \, {\rm M}_\odot$ and redshift 〈z〉 = 0.55. We analyse in parallel a set of zoom-in hydrodynamical simulations from the three hundred project. The SPT-SZ data show two features: (i) a pressure deficit at R/R200m = 1.08 ± 0.09, measured at 3.1σ significance and not observed in the simulations, and; (ii) a sharp decrease in pressure at R/R200m = 4.58 ± 1.24 at 2.0σ significance. The pressure deficit is qualitatively consistent with a shock-induced thermal non-equilibrium between electrons and ions, and the second feature is consistent with accretion shocks seen in previous studies. We split the cluster sample by redshift and mass, and find both features exist in all cases. There are also no significant differences in features along and across the cluster major axis, whose orientation roughly points towards filamentary structure. As a consistency test, we also analyse clusters from the Planck and Atacama Cosmology Telescope Polarimeter surveys and find quantitatively similar features in the pressure profiles. Finally, we compare the accretion shock radius ($R_{\rm sh,\, acc}$) with existing measurements of the splashback radius (Rsp) for SPT-SZ and constrain the lower limit of the ratio, $R_{\rm sh,\, acc}/R_{\rm sp}\gt 2.16 \pm 0.59$.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

PaKman: Scalable Assembly of Large Genomes on Distributed Memory Machines

De novo genome assembly is a fundamental problem in the field of bioinformatics, that aims to assemble the DNA sequence of an unknown genome from numerous short DNA fragments (aka reads) obtained from it. With the advent of high-throughput sequencing technologies, billions of reads can be generated in a matter of hours, necessitating efficient parallelization of the assembly process. While multiple parallel solutions have been proposed in the past, conducting a large- scale assembly at scale remains a challenging problem because of the inherent complexities associated with data movement, and irregular access footprints of memory and I/O operations. In this paper, we present a novel algorithm, called PaKman, to address the problem of performing large-scale genome assemblies on a distributed memory parallel computer. Our approach focuses on improving performance through a combination of novel data structures and algorithmic strategies for reducing the communication and I/O footprint during the assembly process. A key aspect of our algorithm is its graph data structure, which comprises fat nodes (or what we call "macro-nodes") that reduce the communication burden during contig generation. We present an extensive performance and qualitative evaluation of our algorithm, including comparisons to other state-of-the-art parallel assemblers. Our results demonstrate the ability to achieve near-linear speedups on up to 8K cores (tested); outperform state-of-the-art distributed memory and shared memory tools in performance while delivering comparable (if not better) quality; and reduce time to solution significantly. For instance, PaKman is able to complete an assembly of the full human genome in just over a minute on 8K cores.

Ghosh, Priyanka↗
DOE OSTI2017

UMAMI: A Recipe for Generating Meaningful Metrics through Holistic I/O Performance Analysis

I/O efficiency is essential to productivity in scientific computing, especially as many scientific domains become more data-intensive. Many characterization tools have been used to elucidate specific aspects of parallel I/O performance, but analyzing components of complex I/O subsystems in isolation fails to provide insight into critical questions: how do the I/O components interact, what are reasonable expectations for application performance, and what are the underlying causes of I/O performance problems? To address these questions while capitalizing on existing component-level characterization tools, we propose an approach that combines on-demand, modular synthesis of I/O characterization data into a unified monitoring and metrics interface (UMAMI) to provide a normalized, holistic view of I/O behavior. We evaluate the feasibility of this approach by applying it to a month-long benchmarking study on two distinct largescale computing platforms. We present three case studies that highlight the importance of analyzing application I/O performance in context with both contemporaneous and historical component metrics, and we provide new insights into the factors affecting I/O performance. By demonstrating the generality of our approach, we lay the groundwork for a production-grade framework for holistic I/O analysis.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2016

Real-time microstructure imaging by Laue microdiffraction: A sample application in laser 3D printed Ni-based superalloys

Synchrotron-based Laue microdiffraction has been widely applied to characterize the local crystal structure, orientation, and defects of inhomogeneous polycrystalline solids by raster scanning them under a micro/nano focused polychromatic X-ray probe. In a typical experiment, a large number of Laue diffraction patterns are collected, requiring novel data reduction and analysis approaches, especially for researchers who do not have access to fast parallel computing capabilities. In this article, a novel approach is developed by plotting the distributions of the average recorded intensity and the average filtered intensity of the Laue patterns. Visualization of the characteristic microstructural features is realized in real time during data collection. As an example, this method is applied to image key features such as microcracks, carbides, heat affected zone, and dendrites in a laser assisted 3D printed Ni-based superalloy, at a speed much faster than data collection. Such analytical approach remains valid for a wide range of crystalline solids, and therefore extends the application range of the Laue microdiffraction technique to problems where real-time decision-making during experiment is crucial (for instance time-resolved non-reversible experiments).

36 MATERIALS SCIENCE↗
DOE OSTI2023

Model-parallel Fourier neural operators as learned surrogates for large-scale parametric PDEs

Fourier neural operators (FNOs) are a recently introduced neural network architecture for learning solution operators of partial differential equations (PDEs), which have been shown to perform significantly better than comparable deep learning approaches. Once trained, FNOs can achieve speed-ups of multiple orders of magnitude over conventional numerical PDE solvers. However, due to the high dimensionality of their input data and network weights, FNOs have so far only been applied to two-dimensional or small three-dimensional problems. To remove this limited problem-size barrier, we propose a model-parallel version of FNOs based on domain-decomposition of both the input data and network weights. Here, we demonstrate that our model-parallel FNO is able to predict time-varying PDE solutions of over 2.6 billion variables on Perlmutter using up to 512 A100 GPUs and show an example of training a distributed FNO on the Azure cloud for simulating multiphase CO 2 dynamics in the Earth’s subsurface.

58 GEOSCIENCES↗
DOE OSTI2016

Systematic and quantitative assessment of the effect of chronic kidney disease on CYP2D6 and CYP3A4/5

Recent reviews suggest that chronic kidney disease (CKD) can affect the pharmacokinetics of nonrenally eliminated drugs, but the impact of CKD on individual elimination pathways has not been systematically evaluated. In this study we developed a comprehensive dataset of the effect of CKD on the pharmacokinetics of CYP2D6- and CYP3A4/5-metabolized drugs. Drugs for evaluation were selected based on clinical drug–drug interaction (CYP3A4/5 and CYP2D6) and pharmacogenetic (CYP2D6) studies. Information from dedicated CKD studies was available for 13 and 18 of the CYP2D6 and CYP3A4/5 model drugs, respectively. Analysis of these data suggested that CYP2D6-mediated clearance is generally decreased in parallel with the severity of CKD. There was no apparent relationship between the severity of CKD and CYP3A4/5-mediated clearance. The observed elimination-route dependency in CKD effects between CYP2D6 and CYP3A4/5 may inform the need to conduct clinical CKD studies with nonrenally eliminated drugs for optimal use of drugs in patients with CKD.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2016

Effect of PbI{sub 2} deposition rate on two-step PVD/CVD all-vacuum prepared perovskite

In this work we fabricate all-vacuum processed methyl ammonium lead halide perovskite by a sequence of physical vapour deposition of PbI{sub 2} and chemical vapour deposition (CVD) of CH{sub 3}NH{sub 3}I under a static atmosphere. We demonstrate that for higher deposition rate the (001) planes of PbI{sub 2} film show a higher degree of alignment parallel to the sample's surface. From X-ray diffraction data of the resulted perovskite film we derive that the intercalation rate of CH{sub 3}NH{sub 3}I is fostered for PbI{sub 2} films with higher degree of (001) planes alignment. The stoichiometry of the produced perovskite film is also studied by Hard X-ray photoelectron spectroscopy measurements. Complete all-vacuum perovskite solar cells were fabricated on glass/ITO substrates coated by an ultra-thin (5 nm) Zn-phthalocyanine film as hole selective layer. A dependence of residual PbI{sub 2} on the solar cells performance is displayed, while photovoltaic devices with efficiency up to η=11.6% were achieved. - Graphical abstract: A two-step PVD/CVD processed perovskite film with the CVD intercalation rate of CH{sub 3}NCH{sub 3} molecules been fostered by increasing the PVD rate of PbI{sub 2} and prolonging the CVD time. - Highlights: • A simple PVD/CVD process for perovskite film production. • Increased PVD rate yields better alignment of the PbI{sub 2} (001) crystallite planes. • CH{sub 3}NH{sub 3}I intercalation process fostered by increased PbI{sub 2} PVD rate. • Stoichiometric CH{sub 3}NH{sub 3}PbI{sub 3} suitable as absorber in photovoltaic applications • Reduced PbI{sub 2} residue at the bottom of CH{sub 3}NH{sub 3}PbI{sub 3} improves device performance.

36 MATERIALS SCIENCE↗
DOE OSTI2018

Notified access in coarray-based hydrodynamics applications on many-core architectures: Design and performance

With the increasing availability of the Remote Direct Memory Access (RDMA) support in computer networks, the so called Partitioned Global Address Space (PGAS) model has evolved in the last few years. Although there are several cases where a PGAS approach can easily solve difficult message passing situations, like in particle tracking and adaptive mesh refinement applications, the producer-consumer pattern, usually adopted in task-based parallelism, can only be implemented inefficiently because of the separation between data transfer and synchronization (which is usually unified in message passing programming models). Here, in this paper, we provide two contributions: (1) we propose an extension for the Fortran language that provides the concept of Notified Access by associating regular coarray variables with event variables. (2) We demonstrate that the MPI extension proposed by foMPI for Notified Access can be used effectively to implement the same concept in a PGAS run-time library like OpenCoarrays. Moreover, for a hydrodynamics mini-application, we found that Fortran 2018 events perform always better than Fortran 2008 sync statements on many-core processors. We finally show how the proposed Notified Access can improve the performance even more.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Novel approaches in hadron spectroscopy

The last two decades have witnessed the discovery of a myriad of new and unexpected hadrons. The future holds more surprises for us, thanks to new-generation experiments. Understanding the signals and determining the properties of the states requires a parallel theoretical effort. To make full use of available and forthcoming data, a careful amplitude modeling is required, together with a sound treatment of the statistical uncertainties, and a systematic survey of the model dependencies. We review the contributions made by the Joint Physics Analysis Center to the field of hadron spectroscopy.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2017

Cluster analysis of accelerated molecular dynamics simulations: A case study of the decahedron to icosahedron transition in Pt nanoparticles

Modern molecular-dynamics-based techniques are extremely powerful to investigate the dynamical evolution of materials. With the increase in sophistication of the simulation techniques and the ubiquity of massively parallel computing platforms, atomistic simulations now generate very large amounts of data, which has to be carefully analyzed in order to reveal key features of the underlying trajectories, including the nature and characteristics of the relevant reaction pathways. We show that clustering algorithms, such as the Perron Cluster Cluster Analysis, can provide reduced representations that greatly facilitate the interpretation of complex trajectories. To illustrate this point, clustering tools are used to identify the key kinetic steps in complex Accelerated Molecular Dynamics trajectories exhibiting shape uctuations in Pt nanoclusters. This analysis provides an easily-interpretable coarse representation of the reaction pathways in terms of a handful of clusters, in contrast to the raw trajectory that contains thousands of unique states and tens of thousands of transitions.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2022

Phased plan for the implementation of the time-resolving magnetic recoil spectrometer on the National Ignition Facility (NIF)

The time-resolving magnetic recoil spectrometer (MRSt) is a transformative diagnostic that will be used to measure the time-resolved neutron spectrum from an inertial confinement fusion implosion at the National Ignition Facility (NIF). It uses a CD foil on the outside of the hohlraum to convert fusion neutrons to recoil deuterons. An ion-optical system positioned outside the NIF target chamber energy-disperses and focuses forward-scattered deuterons. A pulse-dilation drift tube (PDDT) subsequently dilates, un-skews, and detects the signal. While the foil and ion-optical system have been designed, the PDDT requires more development before it can be implemented. Therefore, a phased plan is presented that first uses the foil and ion-optical systems with detectors that can be implemented immediately—namely CR-39 and hDISC streak cameras. These detectors will allow the MRSt to be commissioned in an intermediate stage and begin collecting data on a reduced timescale, while the PDDT is developed in parallel. A CR-39 detector will be used in phase 1 for the measurement of the time-integrated neutron spectra with excellent energy-resolution, necessary for the energy calibration of the system. Streak cameras will be used in phase 2 for measurement of the time-resolved spectrum with limited spectral coverage, which is sufficient to diagnose the time-resolved ion temperature. Simulations are presented that predict the performance of the streak camera detector, indicating that it will achieve excellent burn history measurements at current yields, and good time-resolved ion-temperature measurements at yields above 3 × 10 17 . The PDDT will be used for optimal efficiency and resolution in phase 3.

47 OTHER INSTRUMENTATION↗
DOE OSTI2016

On the Impact of Widening Vector Registers on Sequence Alignment

Vector extensions, such as SSE, have been part of the x86 since the 1990s, with applications in graphics, signal processing, and scientific applications. Although many algorithms and applications can naturally benefit from automatic vectorization techniques, there are still many that are difficult to vectorize due to their dependence on irregular data structures, dense branch operations, or data dependencies. Sequence alignment, one of the most widely used operations in bioinformatics workflows, has a computational footprint that features complex data dependencies. In this paper, we demonstrate that the trend of widening vector registers adversely affects the state-of-the-art sequence alignment algorithm based on striped data layouts. We present a practically efficient SIMD implementation of a parallel scan based sequence alignment algorithm that can better exploit wider SIMD units. We conduct comprehensive workload and use case analyses to characterize the relative behavior of the striped and scan approaches and identify the best choice of algorithm based on input length and SIMD width.

Needleman-Wunsch↗
DOE OSTI2022

Scalable Deep Learning-Based Microarchitecture Simulation on GPUs

Cycle-accurate microarchitecture simulators are essential tools for designers to architect, estimate, optimize, and manufacture new processors that meet specific design expectations. However, conventional simulators based on discrete-event methods often require an exceedingly long time-to-solution for the simulation of applications and architectures at full complexity and scale. Given the excitement around wielding the machine learning (ML) hammer to tackle various architecture problems, there have been attempts to employ ML to perform architecture simulations, such as Ithemal and SimNet. However, the direct application of existing ML approaches to architecture simulation may be even slower due to overwhelming memory traffic and stringent sequential computation logic. This work proposes the first graphics processing unit (GPU)-based microarchitecture simulator that fully unleashes the potential of GPUs to accelerate state-of-the-art ML-based simulators. First, considering the application traces are loaded from central processing unit (CPU) to GPU for simulation, we introduce various designs to reduce the data movement cost between CPUs and GPUs. Second, we propose a parallel simulation paradigm that partitions the application trace into sub-traces to simulate them in parallel with rigorous error analysis and effective error correction mechanisms. Combined, this scalable GPU-based simulator outperforms by orders of magnitude the traditional CPU-based simulators and the state-of-the-art ML-based simulators, i.e., SimNet and Ithemal.

97 MATHEMATICS AND COMPUTING↗