Search NASA⌕ Search

SEARCH · Search NASA

Results for “graphic processing units”

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 775 records · Page 43

Experiences with High-Level Programming Directives for Porting Applications to GPUs. In: Facing the Multicore--Challenge II, Lecture Notes in Computer Science.

HPC systems now exploit GPUs within their compute nodes to accelerate program performance. As a result, high-end application development has become extremely complex at the node level. In addition to restructuring the node code to exploit the cores and specialized devices, the programmer may need to choose a programming model such as OpenMP or CPU threads in conjunction with an accelerator programming model to share and manage the different node resources. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. In order to offset the high development cost of creating CUDA or OpenCL kernels, directives have been proposed for programming accelerator devices, but their implications are not well known. In this paper, we evaluate the state of the art accelerator directives to program several applications kernels, explore transformations to achieve good performance, and examine the expressivity and performance penalty of using high-level directives versus CUDA. We also compare our results to OpenMP implementations to understand the benefits of running the kernels in the accelerator versus CPU cores.

CUDA Implementation↗

Accelerating Machine Learning Inference with GPUs in ProtoDUNE Data Processing

Abstract We study the performance of a cloud-based GPU-accelerated inference server to speed up event reconstruction in neutrino data batch jobs. Using detector data from the ProtoDUNE experiment and employing the standard DUNE grid job submission tools, we attempt to reprocess the data by running several thousand concurrent grid jobs, a rate we expect to be typical of current and future neutrino physics experiments. We process most of the dataset with the GPU version of our processing algorithm and the remainder with the CPU version for timing comparisons. We find that a 100-GPU cloud-based server is able to easily meet the processing demand, and that using the GPU version of the event processing algorithm is two times faster than processing these data with the CPU version when comparing to the newest CPUs in our sample. The amount of data transferred to the inference server during the GPU runs can overwhelm even the highest-bandwidth network switches, however, unless care is taken to observe network facility limits or otherwise distribute the jobs to multiple sites. We discuss the lessons learned from this processing campaign and several avenues for future improvements.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Accelerated application development: The ORNL Titan experience

The use of computational accelerators such as NVIDIA GPUs and Intel Xeon Phi processors is now widespread in the high performance computing community, with many applications delivering impressive performance gains. However, programming these systems for high performance, performance portability and software maintainability has been a challenge. In this paper we discuss experiences porting applications to the Titan system. Titan, which began planning in 2009 and was deployed for general use in 2013, was the first multi-petaflop system based on accelerator hardware. To ready applications for accelerated computing, a preparedness effort was undertaken prior to delivery of Titan. In this paper we report experiences and lessons learned from this process and describe how users are currently making use of computational accelerators on Titan.

97 MATHEMATICS AND COMPUTING↗

Gyrokinetic particle-in-cell optimization on emerging multi- and manycore platforms

The next decade of high-performance computing (HPC) systems will see a rapid evolution and divergence of multi- and manycore architectures as power and cooling constraints limit increases in microprocessor clock speeds. Understanding efficient optimization methodologies on diverse multicore designs in the context of demanding numerical methods is one of the greatest challenges faced today by the HPC community. In this paper, we examine the efficient multicore optimization of GTC, a petascale gyrokinetic toroidal fusion code for studying plasma microturbulence in tokamak devices. For GTC’s key computational components (charge deposition and particle push), we explore efficient parallelization strategies across a broad range of emerging multicore designs, including the recently-released Intel Nehalem-EX, the AMD Opteron Istanbul, and the highly multithreaded Sun UltraSparc T2+. We also present the first study on tuning gyrokinetic particle-in-cell (PIC) algorithms for graphics processors, using the NVIDIA C2050 (Fermi). Our work discusses several novel optimization approaches for gyrokinetic PIC, including mixed-precision computation, particle binning and decomposition strategies, grid replication, SIMDized atomic floating-point operations, and effective GPU texture memory utilization. Overall, we achieve significant performance improvements of 1.3–4.7× on these complex PIC kernels, despite the inherent challenges of data dependency and locality. Finally, our work also points to several architectural and programming features that could significantly enhance PIC performance and productivity on next-generation architectures.

97 MATHEMATICS AND COMPUTING↗

Achieving performance portability in Gaussian basis set density functional theory on accelerator based architectures in NWChemEx

The numerical integration of the exchange–correlation (XC) potential is one of the primary computational bottlenecks in Gaussian basis set Kohn–Sham density functional theory (KS-DFT). To achieve optimal performance and accuracy, care must be taken in this numerical integration to preserve local sparsity as to allow for near linear weak scaling with system size. This leads to an integration scheme with several performance critical kernels which must be hand optimized for each architecture of interest. As the set of available accelerator hardware goes more diverse, a key challenge for developers of KS-DFT software is to maintain performance portability across a wide range of computational architectures. In this article, we examine a modular software design pattern which decouples the implementation details of performance critical kernels from the expression of high-level algorithmic workflows in a device-agnostic language such as C++; thus allowing for developers to target existing and emerging accelerator hardware within a single code base. We consider the efficacy of such a design pattern in the numerical integration of the XC potential by demonstrating its ability to achieve performance portability across a set of accelerator architectures which are representative of those on current and future U.S. Department of Energy Leadership Computing Facilities.

97 MATHEMATICS AND COMPUTING↗

In Silico Discovery of Multistep Chemistry Initiated by a Conical Intersection: The Challenging Case of Donor–Acceptor Stenhouse Adducts

Detailed mechanistic understanding of multistep chemical reactions triggered by internal conversion via a conical intersection is a challenging task that emphasizes limitations in theoretical and experimental techniques. We present a discovery-based, hypothesis-free computational approach based on first-principles molecular dynamics to discover and refine the switching mechanism of donor–acceptor Stenhouse adducts (DASAs). We simulate the photochemical experiment in silico, following the “hot” ground state dynamics for 10 ps after photoexcitation. Using state-of-the-art graphical processing units-enabled electronic structure calculations we performed in total ~2 ns of nonadiabatic ab initio molecular dynamics discovering (a) critical intermediates that are involved in the open-to-closed transformation, (b) several competing pathways which lower the overall switching yield, and (c) key elements for future design strategies. Finally, our dynamics describe the natural evolution of both the nuclear and electronic degrees of freedom that govern the interconversion between DASA ground-state intermediates, exposing significant elements for future design strategies of molecular switches.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

The FastEddy® Resident-GPU Accelerated Large-Eddy Simulation Framework: Moist Dynamics Extension, Validation and Sensitivities of Modeling Non-Precipitating Shallow Cumulus Clouds

Herein we describe the moist dynamics formulation implemented within the graphics processing unit-resident large-eddy simulation FastEddy® model, which includes a simple saturation adjustment scheme for condensation and evaporation processes. Two LES model intercomparison exercises for non-precipitating shallow cumulus clouds are simulated in order to validate this model extension, including a static forcing and a time-dependent forcing case. Overall, we find our dynamical, thermodynamical and microphysical quantities, along with turbulence variability and fluxes, to be commensurate with the corresponding model intercomparison results. In addition, sensitivities to specific model settings are investigated. Among these settings, it is shown that boundary layer and cloud layer structure and characteristics are sensitive to use of higher-order advection schemes impacting the vertical distribution of cloud content and associated turbulence statistics. Increasing the timescale of the saturation scheme leads to enhanced liquid water presence and decreases vertical velocity variance within the cloud deck. In some cases, these sensitivities agree with the model-to-model variability reported in the intercomparison exercises, highlighting the important role of specific model implementation choices in the context of shallow cumulus convection simulations. These analyses and findings also provide the basis for future extensions and applications of FastEddy® for modeling moist convection and precipitation scenarios.

54 ENVIRONMENTAL SCIENCES↗

Reduced scaling extended multi-state CASPT2 (XMS-CASPT2) using supporting subspaces and tensor hyper-contraction

We present a reduced scaling formulation of the extended multi-state CASPT2 (XMS-CASPT2) method, which is based on our recently developed state-specific CASPT2 (SS-CASPT2) formulation using supporting subspaces and tensor hyper-contraction. By using these two techniques, the off-diagonal elements of the effective Hamiltonian can be computed with only O(N 3 ) operations and O(N 2 ) memory, where N is the number of basis functions. Furthermore, this limits the overall computational scaling to O(N 4 ) operations and O(N 2 ) memory. Thus, excited states can now be obtained at the same reduced (relative to previous algorithms) scaling we achieved for SS-CASPT2.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Simulations of the shock-driven Kelvin–Helmholtz instability in inclined gas curtains

In this paper, we present simulation results for the two-dimensional, shock-driven Kelvin–Helmholtz instability. Simulations are performed with a Mach 2.0 shock propagating through a finite-thickness curtain of gas inclined at an angle α0=30° with respect to the shock plane. After the passage of the shock, the gas curtain is accelerated along its axis. A perturbation develops due to shock reflection near the lower wall, and a Kelvin–Helmholtz instability forms near the vertical center of the curtain. This is the first known numerical reproduction of these phenomena that have previously been observed in experiments with an inclined cylindrical gas column. The effects of varying Mach number and column width were explored in detail to complement experimental data. Additionally, the dependence of the Kelvin–Helmholtz wavelength on Mach number closely matches the relationship observed in experiments. This supports the notion that the observed instability is effectively two-dimensional and inviscid (like classical Kelvin–Helmholtz). The growth rate of the perturbations in the gas curtain was also found to be similar for different Mach numbers. The perturbation at the curtain foot, previously unreported in experiments, was found to have a similar relationship to Mach number as the Kelvin–Helmholtz instability. Both perturbation wavelengths are found to be proportional to layer width. Simulations were performed with the fast interfaces and transport in the atmosphere, an exascale ready, graphics processing unit-accelerated compressible flow solver developed at the University of New Mexico.

42 ENGINEERING↗

Cross-correlation image analysis for real-time single particle tracking

Accurately measuring the translations of objects between images is essential in many fields, including biology, medicine, chemistry, and physics. One important application is tracking one or more particles by measuring their apparent displacements in a series of images. Popular methods, such as the center of mass, often require idealized scenarios to reach the shot noise limit of particle tracking and, therefore, are not generally applicable to multiple image types. More general methods, such as maximum likelihood estimation, reliably approach the shot noise limit, but are too computationally intense for use in real-time applications. These limitations are significant, as real-time, shot-noise-limited particle tracking is of paramount importance for feedback control systems. To fill this gap, we introduce a new cross-correlation-based algorithm that approaches shot-noise-limited displacement detection and a graphics processing unit-based implementation for real-time image analysis of a single particle.

Instruments & Instrumentation↗

Accuracy, transferability, and computational efficiency of interatomic potentials for simulations of carbon under extreme conditions

Large-scale atomistic molecular dynamics (MD) simulations provide an exceptional opportunity to advance the fundamental understanding of carbon under extreme conditions of high pressures and temperatures. However, the fidelity of these simulations depends heavily on the accuracy of classical interatomic potentials governing the dynamics of many-atom systems. Here, this study critically assesses several popular empirical potentials for carbon, as well as machine learning interatomic potentials (MLIPs), in their ability to simulate a range of physical properties at high pressures and temperatures, including the diamond equation of state, its melting line, shock Hugoniot, uniaxial compressions, and the structure of liquid carbon. Empirical potentials fail to accurately predict the behavior of carbon under high pressure–temperature conditions. In contrast, MLIPs demonstrate quantum accuracy, with Spectral Neighbor Analysis Potential (SNAP) and atomic cluster expansion (ACE) being the most accurate in reproducing the density functional theory results. ACE displays remarkable transferability despite not being specifically trained for extreme conditions. Furthermore, ACE and SNAP exhibit superior computational performance on graphics processing unit-based systems in billion atom MD simulations, with SNAP emerging as the fastest. In addition to offering practical guidance in selecting an interatomic potential with a fine balance of accuracy, transferability, and computational efficiency, this work also highlights transformative opportunities for groundbreaking scientific discoveries facilitated by quantum-accurate MD simulations with MLIPs on emerging exascale supercomputers.

36 MATERIALS SCIENCE↗

Calibrating a finite-strain phase-field model of fracture for bonded granular materials with uncertainty quantification

To study the mechanical behavior of mock high explosives, an experimental and simulation program was developed to calibrate, with quantified uncertainty, a material model of the bonded granular material Idoxuridine and nitroplasticized Estane-5703. This paper reports on the efficacy of such a framework as a generalizable methodology for calibrating material models against experimental data with uncertainty quantification. Additionally, this paper studies the effect of two manufacturing temperatures and three initial granular configurations on the unconfined compressive behavior of the resulting bonded granular materials. In each of these cases, the same calibration framework was used; in that, hundreds of high-fidelity direct numerical simulations using a new, graphics processing unit-enabled, high-performance finite element method software, Ratel, were run to calibrate a finite-strain phase-field fracture model against experimental data. It was found that manufacturing temperature influenced the elastic response of the mock high explosives, with higher temperatures yielding a stiffer response. By contrast, it was found that the initial configuration of the grains had a negligible impact on the overall behavior of the mock high explosives though it remains possible that local damage accumulation within the specimens could be altered by the initial configurations. Overall, the calibration framework was successful at creating well-calibrated models, showing its usefulness as an engineering and scientific tool.

36 MATERIALS SCIENCE↗

Parallel interior-point solver for block-structured nonlinear programs on SIMD/GPU architectures

Here, we investigate how to port the standard interior-point method to new exascale architectures for block-structured nonlinear programs with state equations. Computationally, we decompose the interior-point algorithm into two successive operations: the evaluation of the derivatives and the solution of the associated Karush-Kuhn-Tucker (KKT) linear system. Our method accelerates both operations using two levels of parallelism. First, we distribute the computations on multiple processes using coarse parallelism. Second, each process uses SIMD/GPU accelerators locally to accelerate the operations using fine-grained parallelism. The KKT system is reduced by eliminating the inequalities and the state variables from the corresponding equations. We demonstrate our method's capability on the supercomputer Polaris, a testbed for the future exascale Aurora system. Each node is equipped with four GPUs, a setup amenable to our two-level approach. Our experiments on the stochastic optimal power flow problem show that the reduction method is 50x faster than the sparse linear solver HSL MA57 running in serial on the CPU, and 6x faster than Pardiso running in parallel on CPU on the same number of processes.

97 MATHEMATICS AND COMPUTING↗

DeepMC: a deep learning method for efficient Monte Carlo beamlet dose calculation by predictive denoising in magnetic resonance-guided radiotherapy

Abstract Emerging magnetic resonance (MR) guided radiotherapy affords significantly improved anatomy visualization and, subsequently, more effective personalized treatment. The new therapy paradigm imposes significant demands on radiation dose calculation quality and speed, creating an unmet need for the acceleration of Monte Carlo (MC) dose calculation. Existing deep learning approaches to denoise the final plan MC dose fail to achieve the accuracy and speed requirements of large-scale beamlet dose calculation in the presence of a strong magnetic field for online adaptive radiotherapy planning. Our deep learning dose calculation method, DeepMC, addresses these needs by predicting low-noise dose from extremely noisy (but fast) MC-simulated dose and anatomical inputs, thus enabling significant acceleration. DeepMC simultaneously reduces MC sampling noise and predicts corrupted dose buildup at tissue-air material interfaces resulting from MR-field induced electron return effects. Here we demonstrate our model’s ability to accelerate dose calculation for daily treatment planning by a factor of 38 over traditional low-noise MC simulation with clinically meaningful accuracy in deliverable dose and treatment delivery parameters. As a post-processing approach, DeepMC provides compounded acceleration of large-scale dose calculation when used alongside established MC acceleration techniques in variance reduction and graphics processing unit-based MC simulation.

Engineering↗

New constraints on warm dark matter from the Lyman- α forest power spectrum

The forest of Lyman-α absorption lines detected in the spectra of distant quasars encodes information on the nature and properties of dark matter and the thermodynamics of diffuse baryonic material. Its main observable—the 1D flux power spectrum (FPS)—should exhibit a suppression on small scales and an enhancement on large scales in warm dark matter (WDM) cosmologies compared to standard Λ⁢CDM. Here, we present an unprecedented suite of 1080 high-resolution cosmological hydrodynamical simulations run with the graphics processing unit-accelerated code cholla to study the evolution of the Lyman-α forest under a wide range of physically motivated gas thermal histories along with different free-streaming lengths of WDM thermal relics in the early Universe. A statistical comparison of synthetic data with the forest FPS measured down to the smallest velocity scales ever probed at redshifts 4.0≲z≲5.2 [E. Boera et al., Revealing reionization with the thermal history of the intergalactic medium: New constraints from the Ly⁢α flux power spectrum, Astrophys. J. 872, 101 (2019)] yields a lower-limit m WDM >3.1 keV (95% C.L.) for the WDM particle mass and constrains the amplitude and spectrum of the photoheating and photoionizing background produced by star-forming galaxies and active galactic nuclei at these redshifts. Interestingly, our Bayesian inference analysis appears to weakly favor WDM models with a peak likelihood value at the thermal relic mass of m WDM =4.5 keV. In conclusion, we find that the suppression of the FPS from free-streaming saturates at k≳0.1 s km -1 because of peculiar velocity smearing, and this saturated suppression combined with a slightly lower gas temperature provides a moderately better fit to the observed small-scale FPS for WDM cosmologies.

79 ASTRONOMY AND ASTROPHYSICS↗

A High-Throughput Solver for Marginalized Graph Kernels on GPU

Here, we present the design and optimization of a solver for efficient and high-throughput computation of the marginalized graph kernel on General Purpose GPUs. The graph kernel is computed using the conjugate gradient method to solve a generalized Laplacian of the tensor product between a pair of graphs. To cope with the large gap between the instruction throughput and the memory bandwidth of the GPUs, our solver forms the graph tensor product on-the-fly without storing it in memory. This is achieved by using threads in a warp cooperatively to stream the adjacency and edge label matrices of individual graphs by small square matrix blocks called tiles, which are then staged in registers and the shared memory for later reuse. Warps across a thread block can further share tiles via the shared memory to increase data reuse. We exploit the sparsity of the graphs hierarchically by storing only non-empty tiles using a coordinate format and nonzero elements within each tile using bitmaps. We propose a new partition-based reordering algorithm for aggregating nonzero elements of the graphs into fewer but denser tiles to further exploit sparsity. We carry out extensive theoretical analyses on the graph tensor product primitives for tiles of various density and evaluate their performance on synthetic and real-world datasets. Our solver delivers three to four orders of magnitude speedup over existing CPU-based solvers such as GraKeL and GraphKernels. The capability of the solver enables kernel-based learning tasks at unprecedented scales.

97 MATHEMATICS AND COMPUTING↗