Search NASA⌕ Search

SEARCH · Search NASA

Results for “hardware accelerator”

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 631 records · Page 35
All sourcesNASA NTRSDOE OSTI
NASA NTRS2004

Accelerated Training at Mach 20: A Brief Communication Submitted from the International Space Station

The performance of complex tasks on the International Space Station (ISS) requires significant preflight crew training commitments and frequent skill and knowledge refreshment. This report documents a recently developed just-in-time training methodology, which integrates preflight hardware familiarization and procedure training with an on-orbit CD-ROM-based skill enhancement. This just-in-time concept was used to support real-time remote expert guidance to complete medical examinations using the ISS Human Research Facility (HRF). An American and Russian ISS crewmember received 2-hours of hands on ultrasound training 8 months prior to the on-orbit ultrasound exam. A CD-ROM-based Onboard Proficiency Enhancement (OPE) interactive multimedia program consisting of memory enhancing tutorials, and skill testing exercises, was completed by the crewmember six days prior to the on-orbit ultrasound exam. The crewmember was then remotely guided through a thoracic, vascular, and echocardiographic examination by ultrasound imaging experts. Results of the CD ROM based OPE session were used to modify the instructions during a complete 35 minute real-time thoracic, cardiac, and carotid/jugular ultrasound study. Following commands from the ground-based expert, the crewmember acquired all target views and images without difficulty. The anatomical content and fidelity of ultrasound video were excellent and adequate for clinical decision-making. Complex ultrasound experiments with expert guidance were performed with high accuracy following limited pre-flight training and CD-ROM-based in-flight review, despite a 2-second communication latency. In-flight application of multimedia proficiency enhancement software, coupled with real-time remote expert guidance, can facilitate the performance of complex demanding tasks.

Foale, C. Michael↗
DOE OSTI2019

Implementation of a Particle Accelerator Beam Dynamics Code on Multi-Node GPUs

Particle accelerators play an important role in a wide range of scientific discoveries and industrial applications. The self-consistent multi-particle simulation based on the particle-in-cell (PIC) method has been used to study charged particle beam dynamics inside those accelerators. However, the PIC simulation is time-consuming and needs to use modern parallel computers for high-resolution applications. In this paper, we implemented a parallel beam dynamics PIC code on multi-node hybrid architecture computers with multiple Graphics Processing Units (GPUs). We used two methods to parallelize the PIC code on multiple GPUs and observed that the replication method is a better choice for moderate problem size and current computer hardware while the domain decomposition method might be a better choice for large problem size and more advanced computer hardware that allows direct communications among multiple GPUs. Using the multi-node hybrid architectures at Oak Ridge Leadership Computing Facility (OLCF), the optimized GPU PIC code achieves a reasonable parallel performance and scales up to 64 GPUs with 16 million particles.

43 PARTICLE ACCELERATORS↗
DOE OSTI2025

Introduction: Neuromorphic Materials

The explosive growth in data collection and the need to process it efficiently, as well as the desire to automate increasingly complex tasks in transportation, medical care, manufacturing, security and many other fields have motivated a growing interest in neuromorphic computing. Unlike the binary, transistorbased ON/OFF logic gates and separate logic and memory functionalities employed in digital computing, neuromorphic computing is inspired by animal brains that use interconnected synapses and neurons to perform processing, storage and transmission of information at the same location, while only consuming ~20 W or less of power. Motivated by the brain’s efficiency, adaptability, self-learning and resiliency qualities, neuromorphic computing can be broadly defined as an approach to processing and storing information using hardware and algorithms inspired by models of biological neural systems. Present research in neuromorphic computing encompasses approaches that vary significantly in their degree of neuro-inspiration, from systems that only incorporate features such as asynchronous, event-driven operation or use crossbar arrays of non-volatile memory (NVM) elements to accelerate deep neural networks (DNNs), to designs that embrace the extreme parallelism, sparsity, reconfigurability, adaptability, complexity and stochasticity observed in nervous systems. The term ‘neuromorphic’ computing is often credited to Carver Mead, who in the 1980s investigated Si-based analog electronics to replicate functions of the animal retina. Earlier important advances in this field include the work of Frank Rosenblatt, who proposed the concept of the perceptron, Bernard Widrow, who used this concept to build one of the first analog neural networks, the Adaline and many other researchers (see ref. 6 for an historical perspective on neuromorphic computing). With the recent increase in the use of artificial intelligence and large language models, and rising concerns over the associated energy costs, interest in neuromorphic hardware has expanded rapidly. According to some estimates, driven largely by the drastic growth in the training use of artificial intelligence (AI) models using the current computing architectures, the energy cost of computing is projected to reach the energy supply worldwide by 2045. Furthermore, while this is not a realistic outcome, it means that, if more efficient computing technologies are not developed -- soon -- the world will soon become one where demand for energy and market constraints limit the continued increase of societal access to AI and cloud services from data centers. Data centers used for training and use of these models consume hundreds of terawatt hours of electricity, already past 4% of the US electricity demand.

Circuits↗
NASA NTRS2002

Life-Extending Control for Aircraft Engines Studied

Current aircraft engine controllers are designed and operated to provide both performance and stability margins. However, the standard method of operation results in significant wear and tear on the engine and negatively affects the on-wing life--the time between cycles when the engine must be physically removed from the aircraft for maintenance. The NASA Glenn Research Center and its industrial and academic partners have been working together toward a new control concept that will include engine life usage as part of the control function. The resulting controller will be able to significantly extend the engine's on-wing life with little or no impact on engine performance and operability. The new controller design will utilize damage models to estimate and mitigate the rate and overall accumulation of damage to critical engine parts. The control methods will also provide a means to assess tradeoffs between performance and structural durability on the basis of mission requirements and remaining engine life. Two life-extending control methodologies were studied to reduce the overall life-cycle cost of aircraft engines. The first methodology is to modify the baseline control logic to reduce the thermomechanical fatigue (TMF) damage of cooled stators during acceleration. To accomplish this, an innovative algorithm limits the low-speed rotor acceleration command when the engine has reached a threshold close to the requested thrust. This algorithm allows a significant reduction in TMF damage with only a very small increase in the rise time to reach the commanded rotor speed. The second methodology is to reduce stress rupture/creep damage to turbine blades and uncooled stators by incorporating an engine damage model into the flight mission. Overall operation cost is reduced by an optimization among the flight time, fuel consumption, and component damages. Recent efforts have focused on applying life-extending control technology to an existing commercial turbine engine, and doing so without modifying the hardware or adding sensors. This approach makes it possible to retrofit existing engines with life-extending control technology by changing only the control software in the full-authority digital engine controller (FADEC). The significant results include demonstrating a 20- to 30-percent reduction in TMF damage to the hot section by developing and implementing smart acceleration logic during takeoff. The tradeoff is an increase, from 5.0 to 5.2 sec, in the time required to reach maximum power from ground idle. On a typical flight profile of a cruise at Mach 0.8 at an altitude of 41,000 ft, and cruise time of 104 min, the optimized system showed that a reduction in cruise speed from Mach 0.8 to 0.79 can achieve an estimated 25-to 35-percent creep/rupture damage reduction in the engine's hot section and a fuel savings of 2.1 percent. The tradeoff is an increase in flight time of 1.3 percent (1.4 min).

Guo, Te-Huei↗
DOE OSTI2021

ArborX

ArborX library tackles a problem of efficiently finding geometric objects that are close in space. Variations of this problem, such as finding the nearest neighbors of a point, or finding all objects within a certain distance, are inherent components of applications in many fields. The data may be large so that solving the problem efficiently may require significant computational resources, such as multiple processors or accelerators such as general purpose GPUs. ArborX' main advantage in its ability to solve large problems efficiently utilizing a combination of distributed and on-node parallelism. ArborX can be run efficiently on a wide variety of hardware, including GPUs from different vendors, which distinguishes it from other available libraries which typically choose only few of these. The other advantage is that it supports both types of user problems: spatial problems (useful for intersections and finding objects within certain distance), and nearest neighbor problems. ArborX also supports flexible interface in its interaction with a user. Particularly, it allows a user to call user's own function on a positive match, a functionality not rarely available in other libraries. ArborX implements construction and traversal algorithms using efficient tree structures, such as bounding volume hierarchy (BVH). At its core, it uses linear BVH for its low construction cost and sufficient quality. ArborX is written using C++, and is parallelized using the message passing interface (MPI) for the distributed communication, and the Kokkos library for on-node parallelism. This approach allows ArborX to be run on a wide variety of hardware, from common laptops and desktops to supercomputers while using the same codebase. ArborX also implements several advanced algorithms using geometric search, such as density-based clustering algorithm DBSCAN.

ECP↗
DOE OSTI2025

ArborX 2.0

ArborX library tackles a problem of efficiently finding geometric objects that are close in space. Variations of this problem, such as finding the nearest neighbors of a point, or finding all objects within a certain distance, are inherent components of applications in many fields. The data may be large so that solving the problem efficiently may require significant computational resources, such as multiple processors or accelerators such as general purpose GPUs. ArborX' main advantage in its ability to solve large problems efficiently utilizing a combination of distributed and on-node parallelism. ArborX can be run efficiently on a wide variety of hardware, including GPUs from different vendors, which distinguishes it from other available libraries which typically choose only few of these. The other advantage is that it supports both types of user problems: spatial problems (useful for intersections and finding objects within certain distance), and nearest neighbor problems. ArborX also supports flexible interface in its interaction with a user. Particularly, it allows a user to call user's own function on a positive match, a functionality not rarely available in other libraries. ArborX implements construction and traversal algorithms using efficient tree structures, such as bounding volume hierarchy (BVH). At its core, ArborX uses linear BVH for its low construction cost and sufficient quality. ArborX implements both spatial and nearest-neighbor traversal algorithms. ArborX also provides several clustering algorithms (minimum spanning tree, DBSCAN, HDBSCAN*), interpolation using minimum least squares and ray tracing. ArborX is written using C++, and is parallelized using the message passing interface (MPI) for the distributed communication, and the Kokkos library for on-node parallelism. This approach allows ArborX to be run on a wide variety of hardware, from common laptops and desktops to supercomputers while using the same codebase.

Prokopenko, Andrey [Oak Ridge National Laboratory ↗
DOE OSTI2025

SpacelyProject/spacely-docs

Spacely is an open-source framework for the post-silicon validation of analog, digital, and mixed-signal ASICs (Application-Specific Integrated Circuits) which maximizes the reuse of hardware and software, reducing the time taken to achieve meaningful test results. Spacely specifically addresses the needs of small, flexible ASIC design teams commonly found in academia or research institutions. Spacely was originally developed at Fermi National Accelerator Laboratory.

Quinn, Adam [Fermi National Accelerator Laboratory↗
DOE OSTI2018

WBS STPR 04 Milestone 4 Report

This report documents the completion of milestone STPRO4-4 Kokkos back-ends research, collaborations, development, optimization, and documentation. The Kokkos team updated its existing backend to support the software stack and hardware of DOE's Sierra, Summit and Astra machines. They also collaborated with ECP PathForward vendors on developing backends for possible exa-scale architectures. Furthermore, the team ramped up its engagement with the ISO/C++ committee to accelerate the adoption of features important for the HPC community into the C++ standard.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

WBS STPR 04 Milestone 4 Report

This report documents the completion of milestone STPRO4-4 Kokkos back-ends research, collaborations, development, optimization, and documentation. The Kokkos team updated its existing backend to support the software stack and hardware of DOE's Sierra, Summit and Astra machines. They also collaborated with ECP PathForward vendors on developing backends for possible exa-scale architectures. Furthermore, the team ramped up its engagement with the ISO/C++ committee to accelerate the adoption of features important for the HPC community into the C++ standard.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1994

A method to determine the kinematics of the lower limbs of a subject pedaling a bicycle using encoders and accelerometers

The goal of this research was to determine kinematic parameters of the lower limbs of a subject pedaling a bicycle. An existing measurement system was used as the basis to develop the model to determine position and acceleration of the limbs. The system consists of an ergometer instrumented to provide position of the pedal (foot), accelerometers to be attached to the lower limbs to measure accelerations, a recorder used for filtering, and a computer instrumented with an A/D board and a decoder board. The system is designed to read and record data from accelerometers and encoders. Software has been developed for data collection, analysis and presentation. Based on the measurement system, a two dimensional analytical model has been developed to determine configuration (position, orientation) and kinematics (velocities, accelerations). The model has been implemented in software and verified by simulation. An error analysis to determine the system's accuracy shows that the expected error is well within the specifications of practical applications. When the physical hardware is completed, NASA researchers hope to use the system developed to determine forces exerted by muscles and forces at articulations. This data will be useful in the development of countermeasures to minimize bone loss experienced by astronauts in microgravity conditions.

Liu, Shih-Ching↗
DOE OSTI2018

RF Controls for High-Q Cavities for the LCLS-II

The SLAC National Accelerator Laboratory is building LCLS-II, a new 4 GeV CW superconducting (SCRF) Linac as a major upgrade of the existing LCLS. The LCLS-II Low-Level Radio Frequency (LLRF) collaboration is a multi-lab effort within the Department of Energy (DOE) accelerator complex. The necessity of high longitudinal beam stability of LCLS-II imposes tight amplitude and phase stability requirements on the LLRF system (up to 0.01% in amplitude and 0.01° in phase RMS). This is the first time such requirements are expected of superconducting cavities operating in continuous-wave (CW) mode. Initial measurements on the Cryomodule test stands at partner labs have shown that the early production units are able to meet the extrapolated hardware requirements to achieve such levels of performance. A large effort is currently underway for system integration, Experimental Physics and Industrial Control System (EPICS) controls, transfer of knowledge from the partner labs to SLAC and the production and testing of 76 racks of LLRF equipment.

43 PARTICLE ACCELERATORS↗
DOE OSTI2023

Interactive Quantum Chemistry Enabled by Machine Learning, Graphical Processing Units, and Cloud Computing

Modern quantum chemistry algorithms are increasingly able to accurately predict molecular properties that are useful for chemists in research and education. Despite this progress, performing such calculations is currently unattainable to the wider chemistry community, as they often require domain expertise, computer programming skills, and powerful computer hardware. In this review, we outline methods to eliminate these barriers using cutting-edge technologies. We discuss the ingredients needed to create accessible platforms that can compute quantum chemistry properties in real time, including graphical processing units–accelerated quantum chemistry in the cloud, artificial intelligence–driven natural molecule input methods, and extended reality visualization. We end by highlighting a series of exciting applications that assemble these components to create uniquely interactive platforms for computing and visualizing spectra, 3D structures, molecular orbitals, and many other chemical properties.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Understanding the Design Space of Sparse/Dense Multiphase Dataflows for Mapping Graph Neural Networks on Spatial Accelerators

Graph Neural Networks (GNNs) have garnered a lot of recent interest because of their success in learning representations from graph-structured data across several critical applications in cloud and HPC. Owing to their unique compute and memory characteristics that come from an interplay between dense and sparse phases of computations, the emergence of reconfigurable dataflow (aka spatial) accelerators offers promise for acceleration by mapping optimized dataflows (i.e., computation order and parallelism) for both phases. The goal of this work is to characterize and understand the design-space of dataflow choices for running GNNs on spatial accelerators in order for the compilers to optimize the dataflow based on the workload. Specifically, we propose a taxonomy to describe all possible choices for mapping the dense and sparse phases of GNNs spatially and temporally over a spatial accelerator, capturing both the intra-phase dataflow and the inter-phase (pipelined) dataflow. Using this taxonomy, we do deep-dives into the cost and benefits of several dataflows and perform case studies on implications of hardware parameters for dataflows and value of flexibility to support pipelined execution.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1990

Interactive real time flow simulations

An interactive real time flow simulation technique is developed for an unsteady channel flow. A finite-volume algorithm in conjunction with a Runge-Kutta time stepping scheme was developed for two-dimensional Euler equations. A global time step was used to accelerate convergence of steady-state calculations. A raster image generation routine was developed for high speed image transmission which allows the user to have direct interaction with the solution development. In addition to theory and results, the hardware and software requirements are discussed.

Sadrehaghighi, I.↗
NASA NTRS1994

Hypervelocity Impact Test Facility: A gun for hire

An affordable technique has been developed to duplicate the types of impacts observed on spacecraft, including the Shuttle, by use of a certified Hypervelocity Impact Facility (HIF) which propels particulates using capacitor driven electric gun techniques. The fully operational facility provides a flux of particles in the 10-100 micron diameter range with a velocity distribution covering the space debris and interplanetary dust particle environment. HIF measurements of particle size, composition, impact angle and velocity distribution indicate that such parameters can be controlled in a specified, tailored test designed for or by the user. Unique diagnostics enable researchers to fully describe the impact for evaluating the 'targets' under full power or load. Users regularly evaluate space hardware, including solar cells, coatings, and materials, exposing selected portions of space-qualified items to a wide range of impact events and environmental conditions. Benefits include corroboration of data obtained from impact events, flight simulation of designs, accelerated aging of systems, and development of manufacturing techniques.

Johnson, Calvin R.↗
NASA NTRS2011

Phaeton Mast Dynamics: On-Orbit Characterization of Deployable Masts

The PMD instrument is a set of three custom-designed triaxial accelerometer systems designed specifically to detect and characterize the modal dynamics of deployable masts in orbit. The instrument was designed and built as a payload for the NuSTAR spacecraft, but it is now sponsored by the Air Force Research Laboratory's DSX project. It can detect acceleration levels from 1 micro gram to 0.12g over a frequency range of 0.1Hz to 30Hz, the results of which can support future modeling and designing of deployable mast structures for space. This paper details the hardware architecture and design, calibration test and results, and current status of the PMD instrument.

Phaeton Mast Dynamics (PMD)↗
DOE OSTI2018

The state of Monte Carlo neutron transport on GPU accelerators - A review - 25609

A review of the current state of Monte Carlo neutron transport of Graphical Processing Units (GPUs) has been performed. Up-to-date data has been solicited from researchers in the field and combined with data available in literature to provide the current state of Monte Carlo neutron transport on GPU hardware. Nine implementations of neutron transport on GPUs have been summarized and the performance results have been presented to allow the reader to compare performance. A comparison of the results has shown that researchers obtained similar performance when scaled to the performance of NVIDIA Tesla K40 GPUs. A strong scaling equation will be presented that can be used to describe the GPU performance against different numbers of CPU cores. One implementation of a production Monte Carlo code on GPU hardware, the code Shift, showed that newer NVIDIA Pascal and Volta based GPUs provide a significant performance improvement. (authors)

61 RADIATION PROTECTION AND DOSIMETRY↗
DOE OSTI2021

Co-design of Advanced Architectures for Graph Analytics using Machine Learning

A graph is an excellent way of representing relationships among entities. We can use graph analytics to synthesize and analyze such relational data, and extract relevant features that are useful for various tasks such as machine learning. Considering the crucial role of graph analytics in various domains, it is important and timely to investigate the right hardware configurations that can achieve optimal performance for graph workloads on future high-performance computing systems. Design space exploration studies facilitate the selection of appropriate configurations (e.g. memory) to achieve a desired system performance. Recently, the approach of accelerating graph analytics using persistent non-volatile memory has gained a lot of attention. Traditional system simulators such as Gem5 and NVMain can be used to explore the design space of these advanced memory architectures for graph workloads. However, these simulators are slow in execution thus limiting the efficiency of design space exploration studies. To overcome this challenge, we proposed a machine learning based approach to co-design advanced memory architectures for graph workloads. We tested our approach with DRAM, non-volatile memory, and hybrid memory (DRAM+NVM) using a breadth first search benchmark algorithm. Our results showed the applicability of the proposed machine learning based approach to the co-design of the advanced memory architectures. In this paper, we provide recommendations on selecting advanced memory architectures to achieve desired performance for graph workloads. We also discuss the performances of different machine learning models that were considered in this study.

Kurte, Kuldeep↗