Search NASA⌕ Search

SEARCH · Search NASA

Results for “High efficiency acceleration”

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 1,315 records · Page 73

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multi-threaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0x and 6.8x on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3x over the multi-thread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗

Revisiting Huffman Coding: Toward Extreme Performance on Modern GPU Architectures

Today's high-performance computing (HPC) applications are producing vast volumes of data, which are challenging to store and transfer efficiently during the execution, such that data compression is becoming a critical technique to mitigate the storage burden and data movement cost. Huffman coding is arguably the most efficient Entropy coding algorithm in information theory, such that it could be found as a fundamental step in many modern compression algorithms such as DEFLATE. On the other hand, today's HPC applications are more and more relying on the accelerators such as GPU on supercomputers, while Huffman encoding suffers from low throughput on GPUs, resulting in a significant bottleneck in the entire data processing. In this paper, we propose and implement an efficient Huffman encoding approach based on modern GPU architectures, which addresses two key challenges: (1) how to parallelize the entire Huffman encoding algorithm, including codebook construction, and (2) how to fully utilize the high memory-bandwidth feature of modern GPU architectures. The detailed contribution is fourfold. (1) We develop an efficient parallel codebook construction on GPUs that scales effectively with the number of input symbols. (2) We propose a novel reduction based encoding scheme that can efficiently merge the codewords on GPUs. (3) We optimize the overall GPU performance by leveraging the state-of-the-art CUDA APIs such as Cooperative Groups. (4) We evaluate our Huffman encoder thoroughly using six real-world application datasets on two advanced GPUs and compare with our implemented multithreaded Huffman encoder. Experiments show that our solution can improve the encoding throughput by up to 5.0× and 6.8× on NVIDIA RTX 5000 and V100, respectively, over the state-of-the-art GPU Huffman encoder, and by up to 3.3× over the multithread encoder on two 28-core Xeon Platinum 8280 CPUs.

Tian, Jiannan↗

Undulator radiation of single electrons: coherence length and quantum-optical properties

The aims of the CLARA experiment at the Fermilab Integrable Optics Test Accelerator (IOTA) were to directly measure the coherence length of undulator radiation emitted by a single electron and to test whether the radiation is in a pure classical Glauber coherent state or in a quantum mixture of coherent and Fock states. We used a Mach-Zehnder interferometer (MZI) to study visible radiation generated by 150-MeV electrons circulating in the ring. The relative delay between the two arms of the MZI was adjusted by varying the length of one of them with a resolution of 10 nm. The intensity of the circulating beam spanned several orders of magnitude, down to single electrons. A pair of single-photon avalanche diodes (SPADs) was placed at the output of the MZI arms to detect photocounts with high efficiency and timing resolution. We describe the observed interference patterns and photocount rates as a function of interferometer delay and discuss their implications.

43 PARTICLE ACCELERATORS↗

Convolutional neural network based non-iterative reconstruction for accelerating neutron tomography *

Abstract Neutron computed tomography (NCT), a 3D non-destructive characterization technique, is carried out at nuclear reactor or spallation neutron source-based user facilities. Because neutrons are not severely attenuated by heavy elements and are sensitive to light elements like hydrogen, neutron radiography and computed tomography offer a complementary contrast to x-ray CT conducted at a synchrotron user facility. However, compared to synchrotron x-ray CT, the acquisition time for an NCT scan can be orders of magnitude higher due to lower source flux, low detector efficiency and the need to collect a large number of projection images for a high-quality reconstruction when using conventional algorithms. As a result of the long scan times for NCT, the number and type of experiments that can be conducted at a user facility is severely restricted. Recently, several deep convolutional neural network (DCNN) based algorithms have been introduced in the context of accelerating CT scans that can enable high quality reconstructions from sparse-view data. In this paper, we introduce DCNN algorithms to obtain high-quality reconstructions from sparse-view and low signal-to-noise ratio NCT data-sets thereby enabling accelerated scans. Our method is based on the supervised learning strategy of training a DCNN to map a low-quality reconstruction from sparse-view data to a higher quality reconstruction. Specifically, we evaluate the performance of two popular DCNN architectures—one based on using patches for training and the other on using the full images for training. We observe that both the DCNN architectures offer improvements in performance over classical multi-layer perceptron as well as conventional CT reconstruction algorithms. Our results illustrate that the DCNN can be a powerful tool to obtain high-quality NCT reconstructions from sparse-view data thereby enabling accelerated NCT scans for increasing user-facility throughput or enabling high-resolution time-resolved NCT scans.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Pele: An Exascale-Ready Suite of Combustion Codes

High fidelity simulations of realistic combustion devices are extremely demanding computationally because of the requirements to capture complex fuel chemical decomposition, its intricate interactions with turbulent, often multiphase, flows, and the wide separation of space and time scales between the thin flame and the device boundaries. Software required to carry out such computations tends to be extremely complex, particularly when designed to exploit hardware accelerators, and can be difficult to port and maintain. We present Pele, a performance portable suite of tools for the simulation of combustion systems, including codes to evolve reactive multiphase configurations in the low Mach number and compressible flow regimes, along with a set of inter-compatible post processing and in situ analysis tools. The Pele suite of tools is built on top of the AMReX framework for block-structured adaptive mesh refinement, which provides efficient data structures and algorithms that enable the development of a wide variety of efficient mesh and particle based PDE integration schemes. A hierarchical MPI+X parallelism scheme supports CPU-only and accelerated architectures, where X can be OpenMP, CUDA, and HIP based approaches for intra-node computational work distribution. The algorithms and data structures underlying the Pele simulation and analysis tools are highly scalable and performant across a wide variety of high-performance computing platforms, including DOEs newest exascale-class machines, Frontier and Aurora. The simulation and analysis tools are fully documented and freely distributed as open source via GitHub. We present key algorithmic and software challenges, solution strategies, performance and resulting set of capabilities.

AMReX↗

The Extreme Ultraviolet Normal Incidence Spectrograph (EUNIS)

EUNIS (Extreme Ultraviolet Normal Incidence Spectrograph) is a high-efficiency extreme ultraviolet spectrometer that is expected to fly for the first time in 2004 as a sounding rocket payload. Using two independent optical systems, EUNIS will probe the structure and dynamics of the inner solar corona high spectral resolution in two wavelength regions: 17-21 nm with 3.5 pm resolution and 30-37 nm with 7 pm resolution. The long wavelength channel includes He II 30.4 nm and strong lines from Fe XI-XVI; the short wavelength channel includes strong lines of Fe IX-XIII. Angular resolution of 2 arcsec is maintained along a slit covering a full solar radius. EUNIS will have 100 times the throughput of the highly successful SERTS payloads that have preceded it. There are only two reflections in each optical channel, from the superpolished, off-axis paraboloidal primary and the toroidal grating. Each optical element is coated with a high-efficiency multilayer coating optimized for its spectral bandpass. The detector in each channel is a microchannel plate image intensifier fiber- coupled to three 1K x 1K active pixel sensors. EUNIS will obtain spectra with a cadence as short as 1 sec, allowing unprecedented studies of the physical properties of evolving and transient structures. Diagnostics of wave heating and reconnection wil be studied at heights above 2 solar radii, in the wind acceleration region. The broad spectral coverage and high spectral resolution will provide superior temperature and density diagnostics and will enable underflight calibration of several orbital instruments, including SOHO/CDS and EIT, TRACE, Solar-B/EIS, and STEREO/EUVI. EUNIS is supported by NASA through the Low Cost Access to Space Program in Solar and Heliospheric Physics.

Oegerle, William↗

Preparing an Incompressible-Flow Fluid Dynamics Code for Exascale-Class Wind Energy Simulations: Preprint

The US Department of Energy has identified Exascale-Class wind farm simulation tools as critical to wind energy scientific discovery. A primary objective of the Exawind project is to build high-performance, predictive Computational Fluid Dynamics tools that satisfy these modeling needs. GPU accelerators will serve as the computational thoroughbreds of next generation, Exascale-Class, platforms. Here, we report on our efforts for preparing the Exawind unstructured mesh solver, Nalu-Wind, for Exascale-Class machines. For computing at this scale, a simple port of the incompressible-flow algorithms to GPUs is not sufficient. One needs novel algorithms that are application aware, memory efficient, and optimized for latest generation GPU devices to get high-performance. The result of our efforts are unstructured mesh simulations of wind turbines that use 1/6 the compute resources of Summit supercomputer at Oak Ridge National Lab. In particular, we demonstrate a first-of-its-kind, simulation using Algebraic Multigrid solvers on over 4000 GPUs.

algebraic multigrid↗

A GPU‐Accelerated Generative Adversarial Model for Causal Inference

We develop a GPU-accelerated machine learning generative adversarial model designed to facilitate causal inferences from observational data. Our model's theoretical framework is conceptualized in a manner that is amenable to being operable and scalable for high-performance computing platforms. We leverage GPU acceleration to develop a parallel evolutionary algorithm to achieve large-scale parallel computation of the model within a now widely accessible computing platform. This capability both enhances computational speedup and efficiency and also extends the use of the model to a broader range of substantive research domains while maintaining the underlying theoretical properties of the model.

GPU↗

Electrocatalytic oxidation of hydrothermal liquefaction-derived aqueous phase for on-site wastewater treatment and H 2 production

Electrocatalytic oxidation (ECO) is a promising method for generating molecular hydrogen (H 2 ) while simultaneously treating the aqueous phase (AP) from hydrothermal liquefaction (HTL) of biomass-derived feedstocks such as algae, food waste, sludge, and wood. This study highlights the impact of HTL-AP composition on ECO performance, mainly activity, stability, and efficiency in a batch and flow electrolyzer. We demonstrated current efficiency (CE) for chemical oxygen demand (COD) removal ranges from 14% to 85% in the flow electrolyzer with electrode stability increasing from 20 to ≥2,800 h. Decreasing the applied potential enhances the CE. High ammonium content can accelerate deactivation, yet chloride ions appear to aid oxidation of organic compounds and enhanced anode stability. A preliminary energy and H 2 balance for an HTL sewage sludge plant shows that the HTL-AP contains sufficient COD to produce all the H 2 needs for bio-oil hydrotreating, with a 26% surplus available for other uses.

Electrolysis↗

Catalytic reactions within the cavity of coordination cages

Natural enzymes catalyze reactions in their substrate-binding cavities, exhibiting high specificity and efficiency. In an effort to mimic the structure and functionality of enzymes, discrete coordination cages were designed and synthesized. These self-assembled systems have a variety of confined cavities, which have been applied to accelerate conventional reactions, perform substrate-specific reactions, and manipulate regio- and enantio-selectivity. Many coordination cages or cage–catalyst composites have achieved unprecedented results, outperforming their counterparts in different catalytic reactions. This tutorial review summarizes recent developments of coordination cages across three key approaches to coordination cage catalysis: (1) cavity promoted reactions, (2) embedding of active sites in the structure of the cage, and (3) encapsulation of catalysts within the cage. Special emphasis of the review involves (1) introduction of the structure and property of the coordination cage, (2) discussion of the catalytic pathway mediated by the cage, (3) elucidation of the structure–property relationship between the cage and the designated reaction. This work will summarize the recent progress in supramolecular catalysis and attract more researchers to pursue cavity-promoted reactions using discrete coordination cages.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Fast magnetic field annihilation driven by two laser pulses in underdense plasma

Fast magnetic annihilation is investigated by using 2.5-dimensional particle-in-cell simulations of two parallel ultra-short petawatt laser pulses co-propagating in underdense plasma. The magnetic field generated by the laser pulses annihilates in a current sheet formed between the pulses. Magnetic field energy is converted to an inductive longitudinal electric field, which efficiently accelerates the electrons of the current sheet. This new regime of collisionless relativistic magnetic field annihilation with a timescale of tens of femtoseconds can be extended to near-critical and overdense plasma with the ultra-high intensity femtosecond laser pulses.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

THREE-DIMENSIONAL NON-VACUUM PULSAR OUTER-GAP MODEL: LOCALIZED ACCELERATION ELECTRIC FIELD IN THE HIGHER ALTITUDES

We investigate the particle accelerator that arises in a rotating neutron-star magnetosphere. Simultaneously solving the Poisson equation for the electro-static potential, the Boltzmann equations for relativistic electrons and positrons, and the radiative transfer equation, we demonstrate that the electric field is substantially screened along the magnetic field lines by pairs that are created and separated within the accelerator. As a result, the magnetic-field-aligned electric field is localized in higher altitudes near the light cylinder and efficiently accelerates the positrons created in the lower altitudes outward but does not accelerate the electrons inward. The resulting photon flux becomes predominantly outward, leading to typical double-peak light curves, which are commonly observed from many high-energy pulsars.

79 ASTRONOMY AND ASTROPHYSICS↗

Nonideal Fields Solve the Injection Problem in Relativistic Reconnection

Magnetic reconnection in relativistic plasmas is well established as a fast and efficient particle accelerator, capable of explaining the most dramatic astrophysical flares. With particle-in-cell simulations, we demonstrate the importance of nonideal fields for the early stages (“injection”) of particle acceleration. Most of the particles ending up with high energies (near or above the mean magnetic energy per particle) must have passed through nonideal regions where the assumptions of ideal magnetohydrodynamics are broken (i.e., regions with E > B or nonzero E ∥ = E ∙ B/B), whereas most of the particles that do not experience nonideal fields end up with Lorentz factors of order unity. So, injection by nonideal fields is a necessary prerequisite for further acceleration. Our results have important implications for the origin of nonthermal particles in high-energy astrophysical sources.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

ChemComp: A Compilation Framework for Computing with Chemical Reaction Networks

The acceleration of scientific computation, data analytics, and artificial intelligence is driving a surge in computational requirements. Yet, state-of-the-art high-performance computing systems are approaching physical limitations that impede further significant improvements in energy efficiency. As we move towards post-exascale computing systems, innovative approaches are necessary to overcome this barrier in power consumption. Novel analog and hybrid digital-analog architectures hold promise for enhancing energy efficiency by several orders of magnitude. Biochemical computation stands out among the various solutions being explored due to its potential to enable new classes of devices with immense computational capabilities. These devices can capitalize on the inherent efficacy of biological cells in solving optimization problems and are scalable through increasing reaction system size or vessel capacity, potentially satisfying scientific computing's high-performance requirements. Nonetheless, several theoretical and practical limitations persist, including problem formulation and mapping to chemical reaction networks (CRNs) and implementation of actual CRN devices. In this paper, we propose a framework for biochemical computation using systems chemistry. We present the initial components of our approach: an abstract chemical reaction dialect implemented as a multi-level intermediate representation (MLIR) compiler extension and a pathway to represent mathematical problems with CRNs. To showcase the potential of this approach, we emulate a simplified chemical reservoir device. This work lays the groundwork for leveraging chemistry's computing potential in creating energy-efficient, high-performance computing systems tailored to contemporary computational needs.

artificial intelligence↗

Multigrid acceleration of the isenthalpic form of the compressible flow equations

A numerical method for solving the isenthalpic form of the governing equations for compressible inviscid flows was developed. The method is based on the concept of flux vector splitting in its implicit form and was tested on several demanding configurations. Time marching to steady state was accelerated by the implementation of the multigrid procedure which very effectively increased the rate of convergence. High quality steady-state results were obtained for various test cases and required only short computational times due to the relative efficiency of the basic method.

Melson, N. Duane↗

Multigrid acceleration of the isenthalpic form of the compressible flow equations

A numerical method for solving the isenthalpic form of the governing equations for compressible inviscid flows was developed. The method is based on the concept of flux vector splitting in its implicit form and was tested on several demanding configurations. Time marching to steady state was accelerated by the implementation of the multigrid procedure which very effectively increased the rate of convergence. High quality steady-state results were obtained for various test cases and required only short computational times due to the relative efficiency of the basic method.

Melson, N. Duane↗

A spatial operator algebra for manipulator modeling and control

A recently developed spatial operator algebra for manipulator modeling, control, and trajectory design is discussed. The elements of this algebra are linear operators whose domain and range spaces consist of forces, moments, velocities, and accelerations. The effect of these operators is equivalent to a spatial recursion along the span of a manipulator. Inversion of operators can be efficiently obtained via techniques of recursive filtering and smoothing. The operator algebra provides a high-level framework for describing the dynamic and kinematic behavior of a manipulator and for control and trajectory design algorithms. The interpretation of expressions within the algebraic framework leads to enhanced conceptual and physical understanding of manipulator dynamics and kinematics.

Rodriguez, G.↗

NASA HPCC Technology for Aerospace Analysis and Design

The Computational Aerosciences (CAS) Project is part of NASA's High Performance Computing and Communications Program. Its primary goal is to accelerate the availability of high-performance computing technology to the US aerospace community-thus providing the US aerospace community with key tools necessary to reduce design cycle times and increase fidelity in order to improve safety, efficiency and capability of future aerospace vehicles. A complementary goal is to hasten the emergence of a viable commercial market within the aerospace community for the advantage of the domestic computer hardware and software industry. The CAS Project selects representative aerospace problems (especially design) and uses them to focus efforts on advancing aerospace algorithms and applications, systems software, and computing machinery to demonstrate vast improvements in system performance and capability over the life of the program. Recent demonstrations have served to assess the benefits of possible performance improvements while reducing the risk of adopting high-performance computing technology. This talk will discuss past accomplishments in providing technology to the aerospace community, present efforts, and future goals. For example, the times to do full combustor and compressor simulations (of aircraft engines) have been reduced by factors of 320:1 and 400:1 respectively. While this has enabled new capabilities in engine simulation, the goal of an overnight, dynamic, multi-disciplinary, 3-dimensional simulation of an aircraft engine is still years away and will require new generations of high-end technology.

Schulbach, Catherine H.↗