Engineering Papers⌕ Search

SEARCH · Engineering Papers

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 631 records · Page 35
All sourcesNASA NTRSDOE OSTI
NASA NTRS2010

Integration Process for the Habitat Demonstration Unit

The Habitat Demonstration Unit (HDU) is an experimental exploration habitat technology and architecture test platform designed for analog demonstration activities The HDU project has required a team to integrate a variety of contributions from NASA centers and outside collaborators and poses a challenge in integrating these disparate efforts into a cohesive architecture To complete the development of the HDU from conception in June 2009 to rollout for operations in July 2010, a cohesive integration strategy has been developed to integrate the various systems of HDU and the payloads, such as the Geology Lab, that those systems will support The utilization of interface design standards and uniquely tailored reviews have allowed for an accelerated design process Scheduled activities include early fit-checks and the utilization of a Habitat avionics test bed prior to equipment installation into HDU A coordinated effort to utilize modeling and simulation systems has aided in design and integration concept development Modeling tools have been effective in hardware systems layout, cable routing and length estimation, and human factors analysis Decision processes on the shell development including the assembly sequence and the transportation have been fleshed out early on HDU to maximize the efficiency of both integration and field operations Incremental test operations leading up to an integrated systems test allows for an orderly systems test program The HDU will begin its journey as an emulation of a Pressurized Excursion Module (PEM) for 2010 field testing and then may evolve to a Pressurized Core Module (PCM) for 2011 and later field tests, depending on agency architecture decisions The HDU deployment will vary slightly from current lunar architecture plans to include developmental hardware and software items and additional systems called opportunities for technology demonstration One of the HDU challenges has been designing to be prepared for the integration of presently unanticipated systems Results of the HDU field tests will influence future designs of habitat systems.

Gill, Tracy↗
NASA NTRS2007

Quiet Spike(TradeMark) Build-up Ground Vibration Testing Approach

Flight tests of Gulfstream Aerospace Corporation s Quiet Spike(TradeMark) hardware were recently completed on the NASA Dryden Flight Research Center F-15B airplane. NASA Dryden uses a modified F-15B airplane as a testbed aircraft to cost-effectively fly flight research experiments that are typically mounted underneath the F-15B airplane, along the fuselage centerline. For the Quiet Spike(TradeMark) experiment, however, instead of a centerline mounting, a relatively long forward-pointing boom was attached to the radar bulkhead of the F-15B airplane. The Quiet Spike(TradeMark) experiment is a stepping-stone to airframe structural morphing technologies designed to mitigate the sonic-boom strength of business jets over land. The Quiet Spike(TradeMark) boom is a concept in which an aircraft s noseboom would be extended prior to supersonic acceleration. This morphing effectively lengthens the aircraft, thus reducing the peak sonic-boom amplitude, but is also expected to partition the otherwise strong bow shock into a series of reduced-strength, noncoalescing shocklets. Prior to flying the Quiet Spike(TradeMark) experiment on the F-15B airplane several ground vibration tests were required to understand the Quiet Spike(TradeMark) modal characteristics and coupling effects with the F-15B airplane. However, due to the flight hardware availability and compressed schedule requirements, a "traditional" ground vibration test of the mated F-15B Quiet Spike(TradeMark) ready-for- flight configuration did not leave sufficient time available for the finite element model update and flutter analyses before flight testing. Therefore, a "nontraditional" ground vibration testing approach was taken. This paper provides an overview of each phase of the "nontraditional" ground vibration testing completed for the Quiet Spike(TradeMark) project which includes the test setup details, instrumentation layout, and modal results obtained in support of the structural dynamic modeling and flutter analyses.

Spivey, Natalie D.↗
NASA NTRS1972

Long life assurance study for manned spacecraft long life hardware. Volume 4: Special long life assurance studies

Guidelines for the selection of equipment to be used for manned spacecraft in order to assure a five year maintenance-free service life were developed. A special study was conducted to determine the adequacy of the procedures used to determine the quality and effectiveness of various components. The subjects examined are: (1) temperature cycling for acceptance of electronic assemblies; (2) accelerated testing techniques; (3) electronic part screening techniques; (4) electronic part derating practices; (5) vibration life extension of printed circuit board assemblies; and (6) tolerance funnelling and test requirements.

Source record↗
DOE OSTI2024

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↗
NASA NTRS1994

An examination of anticipated g-jitter on Space Station and its effects on materials processes

This study is concerned with the effects of g-jitter, the residual acceleration aboard spacecraft, on selected classes of materials processes. In particular, the anticipated acceleration environment aboard Space Station Freedom (SSF) and its potential effects are analyzed, but the topic is covered with a sufficient level of generality as to apply to other processes and to other vehicles as well. Some of the key findings of this study include: The present acceleration specifications for SSF are inadequate to assure a quality level low-g environment. The local g vector orientation is an extremely sensitive parameter for certain key processes, but can not be controlled to within the desired tolerance. Therefore, less emphasis should be placed upon achieving a tight control of SSF attitude, but more emphasis should be focused on reducing the overall level of the g-jitter magnitude. Melt-based crystal growth may not be successfully processed in the relatively noisy environment of a large inhabited space structure. Growth from vapor or from solution appears more favorable. A smaller space structure and/or a free flyer can provide better alternatives in terms of g-jitter considerations. A high priority (including budgetary) should be given to coordinated efforts among researchers, SSF designers, and equipment contractors, to develop practical experiment-specific sensitivity requirements. Combined focused numerical simulations and experiments with well-resolved acceleration measurements should be vigorously pursued for developing reliable experiment-specific sensitivity data. Appendices provide an extensive cross-referenced bibliography, a discussion of the merits offered by g-jitter analysis techniques, as well as definitions of relevant nondimensional quantities and a brief description of available accelerometry hardware.

Nelson, Emily S.↗
NASA NTRS1996

STS-78 Space Shuttle Mission Report

The STS-78 Space Shuttle Program Mission Report summarizes the Payload activities as well as the Orbiter, External Tank (ET), Solid Rocket Booster (SRB), Reusable Solid Rocket Motor (RSRM), and the Space Shuttle main engine (SSME) systems performance during the seventy-eighth flight of the Space Shuttle Program, the fifty-third flight since the return-to-flight, and the twentieth flight of the Orbiter Columbia (OV-102). In addition to the Orbiter, the flight vehicle consisted of an ET that was designated ET-79; three SSME's that were designated as serial numbers 2041, 2039, and 2036 in positions 1, 2, and 3, respectively; and two SRB's that were designated BI-081. The RSRM's, designated RSRM-55, were installed in each SRB and the individual RSRM's were designated as 360L055A for the left SRB, and 360L055B for the right SRB. The STS-78 Space Shuttle Program Mission Report fulfills the Space Shuttle Program requirement as documented in NSTS 07700, Volume 7, Appendix E. The requirement stated in that document is that each organizational element supporting the Program will report the results of their hardware (and software) evaluation and mission performance plus identify all related in-flight anomalies. The primary objective of this flight was to successfully perform the planned operations of the Life and Microgravity Spacelab experiments. The secondary objectives of this flight were to complete the operations of the Orbital Acceleration Research Experiment (OARE), Biological Research in Canister Unit-Block II (BRIC), and the Shuttle Amateur Radio Experiment II-Configuration C (SAREX-II). The STS-78 mission was planned as a 16-day, plus one day flight plus two contingency days, which were available for weather avoidance or Orbiter contingency operations. The sequence of events for the STS-78 mission is shown in Table 1, and the Space Shuttle Vehicle Management Office Problem Tracking List is shown in Table 2. The Government Furnished Equipment/Flight Crew Equipment (GFE/FCE) Problem Tracking List is shown in Table 3. The Marshall Space Flight Center (MSFC) Problem Tracking List is shown in Table 4. Appendix A lists the sources of data, both formal and informal, that were used to prepare this report. Appendix B provides the definition of acronyms and abbreviations used throughout the report. All times during the flight are given in Greenwich mean time (G.m.t.) and mission elapsed time (MET).

Fricke, Robert W., Jr.↗
NASA NTRS2009

Bayes Analysis and Reliability Implications of Stress-Rupture Testing a Kevlar/Epoxy COPV Using Temperature and Pressure Acceleration

Composite Overwrapped Pressure Vessels (COPVs) that have survived a long service time under pressure generally must be recertified before service is extended. Flight certification is dependent on the reliability analysis to quantify the risk of stress rupture failure in existing flight vessels. Full certification of this reliability model would require a statistically significant number of lifetime tests to be performed and is impractical given the cost and limited flight hardware for certification testing purposes. One approach to confirm the reliability model is to perform a stress rupture test on a flight COPV. Currently, testing of such a Kevlar49 (Dupont)/epoxy COPV is nearing completion. The present paper focuses on a Bayesian statistical approach to analyze the possible failure time results of this test and to assess the implications in choosing between possible model parameter values that in the past have had significant uncertainty. The key uncertain parameters in this case are the actual fiber stress ratio at operating pressure, and the Weibull shape parameter for lifetime; the former has been uncertain due to ambiguities in interpreting the original and a duplicate burst test. The latter has been uncertain due to major differences between COPVs in the database and the actual COPVs in service. Any information obtained that clarifies and eliminates uncertainty in these parameters will have a major effect on the predicted reliability of the service COPVs going forward. The key result is that the longer the vessel survives, the more likely the more optimistic stress ratio model is correct. At the time of writing, the resulting effect on predicted future reliability is dramatic, increasing it by about one "nine," that is, reducing the predicted probability of failure by an order of magnitude. However, testing one vessel does not change the uncertainty on the Weibull shape parameter for lifetime since testing several vessels would be necessary.

Phoenix, S. Leigh↗
NASA NTRS2015

Multi-Axis Independent Electromechanical Load Control for Docking System Actuation Development and Verification Using dSPACE

The development of highly complex and advanced actuation systems to meet customer demands has accelerated as the use of real-time testing technology expands into multiple markets at Moog. Systems developed for the autonomous docking of human rated spacecraft to the International Space Station (ISS), envelope multi-operational characteristics which place unique constraints on an actuation system. Real-time testing hardware has been used as a platform for incremental testing and development for the linear actuation system which controls initial capture and docking for vehicles visiting the ISS. This presentation will outline the role of dSPACE hardware as a platform for rapid control-algorithm prototyping as well as an Electromechanical Actuator (EMA) system dynamic loading simulator, both conducted at Moog to develop the safety critical Linear Actuator System (LAS) of the NASA Docking System (NDS).

Oesch, Christopher↗
DOE OSTI2023

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↗
DOE OSTI2021

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↗
DOE OSTI2026

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↗
DOE OSTI2020

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↗
NASA NTRS1997

The GSFC Combined Approach of ODC Stockpiling and Tribological Testing to Mitigate the Risks of ODC Elimination

In response to the elimination of production of several Ozone Depleting Chemicals (ODC's) which have been widely used in successful space flight mechanism cleaning and lubricating procedures, GSFC developed and implemented an overall philosophy of mitigating the risks to flight hardware during the transition phase to ODC-free cleaning procedures. The short term leg of the philosophy was the stockpiling of an appropriate amount of ODC solvents such that all short term GSFC missions will be able to stay with or revert to heritage cleaning and lubricating procedures in the face of life issues. The long-term leg of that philosophy was the initiation of a several tier testing program that will deliver increasing amounts of information over the next few years, starting with accelerated lubricant life tests that compare lubricant life on surfaces cleaned with ODC solvents with lubricant life on surfaces cleaned with ODC-free solvents. While tribological testing, mechanism life testing and space-flight experience will ultimately bring us into the 21st century with environmentally friendly means of cleaning long-life precision mechanism components, many satellites will be launched over the next few years before a number of important tribological questions can be answered. In order to prepare for this challenge, the Materials Engineering Branch in cooperation with the Electromechanical Branch launched an intensive review of all ongoing missions. The failure risk was determined for each long-life lubricated mechanism based on a number of parameters, including 4 comparison of flight solvents used to clean the heritage/life test hardware. Also studied was the ability of the mechanism manufacturers to stockpile ODC's based on state laws and company policies. A stockpiling strategy was constructed based on this information and subsequently implemented. This paper provides an overview of the GSFC ODC elimination risk mitigation philosophy as well as a detailed examination of the development of the ODC stockpiling plan.

Predmore, Roamer↗
DOE OSTI2024

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↗
DOE OSTI2025

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↗
NASA NTRS1997

Bleed Hole Flow Phenomena Studied

Boundary-layer bleed is an invaluable tool for controlling the airflow in supersonic aircraft engine inlets. Incoming air is decelerated to subsonic speeds prior to entering the compressor via a series of oblique shocks. The low momentum flow in the boundary layer interacts with these shocks, growing in thickness and, under some conditions, leading to flow separation. To remedy this, bleed holes are strategically located to remove mass from the boundary layer, reducing its thickness and helping to maintain uniform flow to the compressor. The bleed requirements for any inlet design are unique and must be validated by extensive wind tunnel testing to optimize performance and efficiency. To accelerate this process and reduce cost, researchers at the NASA Lewis Research Center initiated an experimental program to study the flow phenomena associated with bleed holes. Knowledge of these flow properties will be incorporated into computational fluid dynamics (CFD) models that will aid engine inlet designers in optimizing bleed configurations before any hardware is fabricated. This ongoing investigation is currently examining two hole geometries, 90 and 20 (both with 5-mm diameters), and various flow features.

Source record↗
NASA NTRS2016

Development of Multi-Physics Dynamics Models for High-Frequency Large-Amplitude Structural Response Simulation

An analytic approach is demonstrated to reveal potential pyroshock-driven dynamic effects causing power losses in the Thermo-Electric (TE) module bars of the Mars Science Laboratory (MSL) Multi-Mission Radioisotope Thermoelectric Generator (MMRTG). This study utilizes high-fidelity finite element analysis with SIERRA/PRESTO codes to estimate wave propagation effects due to large-amplitude suddenly-applied pyro shock loads in the MMRTG. A high fidelity model of the TE module bar was created with approximately 30 million degrees-of-freedom (DOF). First, a quasi-static preload was applied on top of the TE module bar, then transient tri-axial acceleration inputs were simultaneously applied on the preloaded module. The applied input acceleration signals were measured during MMRTG shock qualification tests performed at the Jet Propulsion Laboratory. An explicit finite element solver in the SIERRA/PRESTO computational environment, along with a 3000 processor parallel super -computing framework at NASA AMES, was used for the simulation. The simulation results were investigated both qualitatively and quantitatively. The predicted shock wave propagation results provide detailed structural responses throughout the TE module bar, and key insights into the dynamic response (i.e., loads, displacements, accelerations) of critical internal spring/piston compression systems, TE materials, and internal component interfaces in the MMRTG TE module bar. They also provide confidence on the viability of this high-fidelity modeling scheme to accurately predict shock wave propagation patterns within complex structures. This analytic approach is envisioned for modeling shock sensitive hardware susceptible to intense shock environments positioned near shock separation devices in modern space vehicles and systems.

Shock Waves↗
DOE OSTI2025

Parallel sorting algorithm classification: is manual instrumentation necessary?

Understanding parallel algorithms is crucial for accelerating scientific simulations on complex, distributed memory, high-performance computers. Modern algorithm classification approaches learn semantics directly from source code to differentiate between algorithms, however, accessing source code is not always possible. We can learn about parallel algorithms from observing their performance, as programs running the same algorithms and using the same hardware should exhibit similar performance characteristics. We present an approach to learn algorithm classes from parallel performance data directly in order to classify algorithms without access to the source code. We extend previous work to enable classifying parallel sorting algorithms using automatic instrumentation instead of requiring manual region annotations in the source code. In this work, we design and demonstrate a study for classification of parallel sorting algorithms using parallel performance data collected from automatic instrumentation, and evaluate the performance of our new methodology on classification. We leverage Caliper to collect the performance data, Thicket for our exploratory data analysis (EDA), and PyTorch and Scikit-learn to evaluate the effectiveness of random forests, support vector machines (SVMs), decision trees, neural networks, and logistic regressions on parallel performance data. Additionally, we study noise in parallel performance data, whether the removal of noise and pre-processing of the data is necessary to accurately classify parallel sorting algorithms, and determine the effectiveness of features created from performance data. In conclusion, we demonstrate classification accuracy for these five different models of up to 97.7% across four different parallel algorithm classes.

Algorithm Classification↗