Search NASA⌕ Search

SEARCH · Search NASA

Results for “hardware accelerators”

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 757 records · Page 42

Parallel k-means Clustering of Geospatial Data Sets Using Manycore CPU Architectures

The increasing availability of high-resolution geospatiotemporal data sets from sources such as observatory networks, remote sensing platforms, and computational Earth system models has opened new possibilities for knowledge discovery and mining of weather, climate, ecological, and other geoscientific data sets fused from disparate sources. Many of the standard tools used on individual workstations are impractical for the analysis and synthesis of data sets of this size; however, new algorithmic approaches that can effectively utilize the complex memory hierarchies and the extremely high levels of parallelism available in state-of-the-art high-performance computing platforms can enable such analysis. Here, we describe pKluster, an open-source tool we have developed for accelerated k-means clustering of geospatial and geospatiotemporal data, and discuss algorithmic modifications and code optimizations we have made to enable it to effectively use parallel machines based on novel CPU architectures—such as the Intel “Knights Landing” Xeon Phi and Skylake Xeon processors—with many cores and hardware threads, and employing significant single instruction, multiple data (SIMD) parallelism. We outline some applications of the code in ecology and climate science contexts and present a detailed discussion of the performance of the code for one such application, LiDAR-derived vertical vegetation structure classification.

Mills, Richard T↗

Towards Sustainable Post-Exascale Leadership Computing

As computing systems approach the limits of traditional silicon technology, the diminishing returns in performance per watt present a significant barrier to sustaining growth in HPC. From a large-scale scientific supercomputing facility point of view, we propose a multifaceted strategy toward specialized hardware and architectures that are optimized for energy efficiency in specific applications. We also emphasize the need for integrating energy-aware practices across all levels of HPC, from system design and software development to operational policies. We discuss strategic opportunities such as the adoption of application-specific accelerators, the development of energy-efficient algorithms, and the implementation of data-driven operational analytics. Our goal is to develop a comprehensive roadmap ensuring that future leadership systems at OLCF can meet scientific demands while operating within stringent energy budgets, thereby supporting sustainable computing growth.

Shin, Woong↗

Nonlinear Feature Extraction and Energy Dissipation of Foam/Metal Interfaces

The physical mechanisms of energy dissipation in foam to metal interfaces must be understood in order to develop predictive models of systems with foam packaging common to many aerospace and aeronautical applications. Experimental data was obtained from hardware termed Ministack, which has large, unbonded interfaces held under compressive preload. This setup has a solid aluminum mass placed into two foam cups which are then inserted into an aluminum can and fastened with a known preload. Ministack was tested on a shaker using upward sine sweep base acceleration excitations to estimate the linearized natural frequency and energy dissipation of the first axial mode. The experimental system was disassembled and reassembled before each series of tests in order to observe the effects of the assembly to assembly variability on the dynamics. Additionally, Ministack was subjected to upward and downward sweeps to gain some understanding of the nonlinearities. Finally, Ministack was tested using a transient input, and the ring down was analyzed to find the effective stiffness and damping. There are some important findings in the measured data: there is significant assembly to assembly variability, the order in which the sine sweeps are performed influence s the dynamic response, and the system exhibits nontrivial damping and stiffness nonlinearities that must be accounted for in modeling efforts .

42 ENGINEERING↗

The machine protection system for the R&D energy recovery LINAC

The Machine Protection System (MPS) is a device-safety system that is designed to prevent damage to hardware by generating interlocks, based upon the state of input signals generated by selected sub-systems. It protects all the key machinery in the R&D Project called the Energy Recovery LINAC (ERL) against the high beam current. The MPS is capable of responding to a fault with an interlock signal within several microseconds. The ERL MPS is based on a National Instruments CompactRIO platform, and is programmed by utilizing National Instruments' development environment for a visual programming language. The system also transfers data (interlock status, time of fault, etc.) to the main server. Transferred data is integrated into the pre-existing software architecture which is accessible by the operators. This paper will provide an overview of the hardware used, its configuration and operation, as well as the software written both on the device and the server side.

43 PARTICLE ACCELERATORS↗

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↗

Investigation of Portable Event-Based Monte Carlo Transport Using the NVIDIA Thrust Library

Power consumption considerations are driving future high performance computing platforms toward many-core computing architectures. Los Alamos National Laboratory's Trinity machine, available in 2016, will use both Intel Xeon Haswell processors and Intel Xeon Phi Knights Landing many integrated core (MIC) architecture coprocessors. Lawrence Livermore National Laboratory's Sierra machine, available in 2018, will use an IBM PowerPC architecture along with Nvidia graphics processing unit (GPU) architecture accelerators. These different advanced architectures make the computing landscape in upcoming years complex. Traditional approaches to Monte Carlo transport do not work efficiently on these new computing platforms. MIC architectures require vectorization to operate efficiently, and vectorization is difficult to achieve in Monte Carlo transport. GPU architectures require additional code to explicitly use the hardware, requiring significant code changes or hardware specific branches in the source code. A significant challenge for Monte Carlo transport projects is to simultaneously support within a single source code base efficient simulations for both the current generation of architectures and the different advanced computing architectures. In order to address these challenges, two important changes are typically required: a new algorithmic approach for solving Monte Carlo transport, and explicit use of hardware specific software. In this paper, we describe initial research investigations of an event-based Monte Carlo transport algorithm implemented using the Nvidia Thrust library on a GPU for a Monte Carlo test code. The event-based algorithm targets many-core architectures by increasing SIMD (single instruction multiple data) parallelism, while Thrust potentially provides portable performance by allowing one source code base to compile code targeted for both CPUs and GPUs. We described preliminary investigations of portable event-based Monte Carlo algorithms implemented using the Nvidia Thrust library in a research Monte Carlo test code. We found that an explicit CUDA implementation of an event-based Monte Carlo algorithm performed significantly more efficiently than a Thrust implementation on GPU platforms, most likely as a result of additional flexibility in access to different memory spaces on the GPU. Additionally, we showed that on GPU platforms and at large enough problem sizes the event-based implementations perform more efficiently than the serial history-based implementation running on the host CPU. While investigating this problem, we also discovered that the performance of the event-based algorithm is affected by what tallies are being used. A zonal scalar flux tally requires atomic operations that significantly impacted the performance of the code, in some cases producing slowdowns instead of speedups. We decided to remove the tally in order to focus on the effectiveness of the event-based algorithm. Future work will be required to research more effective ways of handling such tallies. Additionally, we would like to consider new ways for optimizing both the Thrust and CUDA versions, in order to see how much performance we have yet to achieve. The potential trade-off between portability and performance was demonstrated in this investigation. Thrust provides both CPU and GPU versions of the code in one code base, but it does so at a cost. We discovered approximately a factor of 2-5 performance difference between Thrust and CUDA on each GPU platform we tested. Thrust provides a tool to access the GPUs with less effort and specialization, but it does so by giving up fine grained control where extra performance can be found for this application. Future work will be required to determine whether this performance differential between Thrust and CUDA can be reduced for event-based Monte Carlo transport. (authors)

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

TES X-ray Spectrometer at SLAC LCLS-II

We are building a transition edge sensor (TES) X-ray spectrometer for the Linac Coherent Light Source (LCLS-II) at SLAC National Accelerator Laboratory (SLAC) to coincide with new upgrades for this free electron laser facility. This new X-ray spectrometer will have 1000 TES pixels with 0.5 eV energy resolution for soft X-rays below 1 keV. Multiplexing will be done with microwave SQUID resonators and new specialized electronic hardware developed at SLAC. This spectrometer will use a dilution refrigerator to achieve lower operating temperatures than previous TES spectrometers and will be coupled to the liquid jet endstation at LCLS-II. The spectrometer is designed to operate at much higher count rates than previous TES X-ray spectrometers to take advantage of the high repetition rate of the LCLS-II. Science applications will utilize the high photon collection efficiency and throughput, high energy resolution, as well as its ability to simultaneously measure its full calibrated energy range.

47 OTHER INSTRUMENTATION↗

Using MLIR Framework for Codesign of ML Architectures Algorithms and Simulation Tools

MLIR (Multi-Level Intermediate Representation), is an extensible compiler framework that supports high-level data structures and operation constructs. These higher-level code representations are particularly applicable to the artificial intelligence and machine learning (AI/ML) domain, allowing developers to more easily support upcoming heterogeneous AI/ML accelerators and develop flexible domain specific compilers/frameworks with higher-level intermediate representations (IRs) and advanced compiler optimizations. The result of using MLIR within the LLVM compiler framework is expected to yield significant improvement in the quality of generated machine code, which in turn will result in improved performance and hardware efficiency

97 MATHEMATICS AND COMPUTING↗

Event-Driven Neuromorphic Accelerator (Caspian)

Software repository for the Caspian neuromorphic system. In order to deploy small, low-power spiking neural networks at the edge, there needs to be 1) an FPGA design for an event simulator to evaluate these neural networks as well as 2) a high-performance cycle-accurate simulator for training these spiking neural networks. This software package solves these problems by providing the Caspian simulator and the µCaspian hardware design in SystemVerilog.

Mitchell, JohnParker [Microsoft Corporation, Redmo↗

National Synchrotron Light Source II storage ring vacuum systems

The National Synchrotron Light Source II, completed in 2014, is a 3-GeV synchrotron radiation (SR) facility at Brookhaven National Laboratory and has been in steady operation since. With a design electron current of 500 mA and subnanometer radians horizontal emittance, this 792-m circumference storage ring is providing the highest flux and brightness x-ray beam for SR users. The majority of the storage ring vacuum chambers are made of extruded aluminium. Chamber sections are interconnected using low-impedance radiofrequency shielded bellows. SR from the bending magnets is intercepted by water-cooled compact photon absorbers resided in the storage ring chambers. This paper presents the design of the storage ring vacuum system, the fabrication of vacuum chambers and other hardware, the installation, the commissioning, and the continuing beam conditioning of the vacuum systems.

43 PARTICLE ACCELERATORS↗

Computing NLTE Opacities -- Node Level Parallel Calculation

Presentation. The goal: to produce a robust library capable of computing reasonably accurate opacities inline with the assumption of LTE relaxed (non-LTE). Near term: demonstrate acceleration of non-LTE opacity computation. Far term (if funded): connect to application codes with in-line capability and compute opacities. Study science problems. Use efficient algorithms that expose many levels of parallelism and utilize good memory access patterns for use on advanced architectures. Portability to multiple types of hardware including multicore processors, manycore processors such as KNL, GPUs, etc. Easily coupled to radiation hydrodynamics and thermal radiative transfer codes.

Astronomy & Astrophysics(79)↗

A Generic High Bandwidth Data Acquisition Card for Physics Experiments

In high energy physics and nuclear physics experiments particularly the ones based on particle accelerator, the data rate from the detector is usually in the order of Terabytes per second. This high throughput data from detector front-end electronics need to be transmitted to the back-end computing farm for high level event selection and building. A Data Acquisition (DAQ) system with features of high-density, scalable, easily upgradeable is crucial to simplify the readout architecture of whole experiment. This paper will introduce the design of a generic high bandwidth PCIe card which can be used as the important input output card in a scalable DAQ system. It can factorize front-end electronics from data handling, and reduce amount of custom hardware in favor of scalable detectorindependent commercial hardware and software. Besides the 48 channels of bidirectional high speed fiber optical links with frontends, it also supports to synchronize with the experiment timing system, and to fanout the clock and trigger information with a fixed latency to the front-end electronics.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Open-source MFIX-DEM software for gas–solids flows: Part I—Verification studies

With rapid advancements in computer hardware, it is now possible to perform large simulations of granular flows using the Discrete Element Method (DEM). As a result, solids are increasingly treated in a discrete Lagrangian fashion in the gas–solids flow community. In this paper, the open-source MFIX-DEM software is described that can be used for simulating the gas–solids flow using an Eulerian reference frame for the continuum fluid and a Lagrangian discrete framework (Discrete Element Method) for the particles. This method is referred to as the continuum discrete method (CDM) to clearly make a distinction between the ambiguity of using a Lagrangian or Eulerian reference for either continuum or discrete formulations. This freely available CDM code for gas–solids flows can accelerate the research in computational gas–solids flows and establish a baseline that can lead to better closures for the continuum modeling (or traditionally referred to as two fluid model) of gas–solids flows. In this paper, a series of verification cases is employed which tests the different aspects of the code in a systematic fashion by exploring specific physics in gas–solids flows before exercising the fully coupled solution on simple canonical problems. It is critical to have an extensively verified code as the physics is complex with highly-nonlinear coupling, and it is difficult to ascertain the accuracy of the results without rigorous verification. These series of verification tests set the stage not only for rigorous validation studies (performed in part II of this paper) but also serve as a procedure for testing any new developments that couple continuum and discrete formulations for gas–solids flows.

Computational fluid dynamics↗

Toward coherent quantum computation of scattering amplitudes with a measurement-based photonic quantum processor

In recent years, applications of quantum simulation have been developed to study the properties of strongly interacting theories. This has been driven by two factors: on the one hand, needs from theorists to have access to physical observables that are prohibitively difficult to study using classical computing; on the other hand, quantum hardware becoming increasingly reliable and scalable to larger systems. In this work, we discuss the feasibility of using quantum optical simulation for studying scattering observables that are presently inaccessible via lattice QCD and are at the core of the experimental program at Jefferson Laboratory, the future Electron-Ion Collider, and other accelerator facilities. We show that recent progress in measurement-based photonic quantum computing can be leveraged to provide deterministic generation of required exotic gates and implementation in a single photonic quantum processor. Published by the American Physical Society 2024

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Poster — Thur Eve — 37: Respiratory gating with an Elekta flattening filter free photon beam

In cases where surgery is not possible for lung cancer treatment, stereotactic body radiation therapy (SBRT) may be an option. One problem when treating this type of cancer is the motion of the lungs caused by the patient's respiration. It is possible to reduce the impact of this movement with the use of respiratory gating. By combining respiratory gating with a flattening filter free (FFF) photon beam linac, the increased treatment time caused by a reduced beam-on time of respiratory gating methods can be compensated by the inherent increased dose rate of FFF beams. This project's aim is to create hardware and software interfaces allowing free respiration gating on an Elekta Synergy-S linac specially modified to deliver 6 MV FFF photon beams. First, a printed circuit board was created for reading the signal from a Bellows Belt from Philips (a respiration monitor belt) and transmitting an On/Off signal to the accelerator. A software was also developed to visualize patient respiration. Secondly, a FFF model was created with the Pinnacle treatment planning system from Philips. Gamma (Γ) analysis (2%, 2 mm) was used to evaluate model. For fields going from 5.6 × 5.6 to 12 × 12 cm{sup 2}, central axis depth dose model fitting shows an average gamma value of 0.2 and 100% of gamma values remain under the Γ = 1 limit. For smaller fields (0.8 × 0.8 and 1.6 × 1.6 cm{sup 2}), Pinnacle has more trouble trying to fit the measurements, overestimating dose in penumbra and buildup regions.

60 APPLIED LIFE SCIENCES↗

Characterizing the Impact of GPU Power Management on an Exascale System

As GPU-accelerated high-performance computing (HPC) systems approach exascale performance, controlling energy consumption without compromising throughput is essential. Architectures such as the AMD MI250X-based Frontier supercomputer provide runtime mechanisms like frequency and power capping, enabling energy tuning without modifying application code. Although both target energy reduction, they operate via distinct hardware control paths and influence workloads differently. We present a comprehensive evaluation of these strategies on a leadership-class system using diverse HPC proxy applications representative of production workloads. Our study analyzes performance–energy trade-offs across multiple capping levels, node counts (1 and 32), and application profiles. Results show that frequency capping generally achieves higher energy efficiency and scalability, with gains of up to 13.2% without performance loss, while power capping is more effective for single-node runs or bursty GPU utilization. We also provide practical guidelines to help system administrators and users balance energy efficiency and performance in large-scale scientific workloads.

Costa, Mariana [Universidade Federal do Rio Grande↗

Energy Efficient Biomolecular Simulations with FPGA-based Reconfigurable Computing

Reconfigurable computing (RC) is being investigated as a hardware solution for improving time-to-solution for biomolecular simulations. A number of popular molecular dynamics (MD) codes are used to study various aspects of biomolecules. These codes are now capable of simulating nanosecond time-scale trajectories per day on conventional microprocessor-based hardware, but biomolecular processes often occur at the microsecond time-scale or longer. A wide gap exists between the desired and achievable simulation capability; therefore, there is considerable interest in alternative algorithms and hardware for improving the time-to-solution of MD codes. The fine-grain parallelism provided by Field Programmable Gate Arrays (FPGA) combined with their low power consumption make them an attractive solution for improving the performance of MD simulations. In this work, we use an FPGA-based coprocessor to accelerate the compute-intensive calculations of LAMMPS, a popular MD code, achieving up to 5.5 fold speed-up on the non-bonded force computations of the particle mesh Ewald method and up to 2.2 fold speed-up in overall time-to-solution, and potentially an increase by a factor of 9 in power-performance efficiencies for the pair-wise computations. The results presented here provide an example of the multi-faceted benefits to an application in a heterogeneous computing environment.

99 GENERAL AND MISCELLANEOUS↗