Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “latency”

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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
NASA NTRS2004

Latency Requirements for Head-Worn Display S/EVS Applications

NASA s Aviation Safety Program, Synthetic Vision Systems Project is conducting research in advanced flight deck concepts, such as Synthetic/Enhanced Vision Systems (S/EVS), for commercial and business aircraft. An emerging thrust in this activity is the development of spatially-integrated, large field-of-regard information display systems. Head-worn or helmet-mounted display systems are being proposed as one method in which to meet this objective. System delays or latencies inherent to spatially-integrated, head-worn displays critically influence the display utility, usability, and acceptability. Research results from three different, yet similar technical areas flight control, flight simulation, and virtual reality are collectively assembled in this paper to create a global perspective of delay or latency effects in head-worn or helmet-mounted display systems. Consistent definitions and measurement techniques are proposed herein for universal application and latency requirements for Head-Worn Display S/EVS applications are drafted. Future research areas are defined.

Bailey, Randall E.↗
NASA NTRS2014

Low-latency Science Exploration of Planetary Bodies: a Demonstration Using ISS in Support of Mars Human Exploration

We summarize a proposed experiment to use the International Space Station to formally examine the application and validation of low-latency telepresence for surface exploration from space as an alternative, precursor, or potentially as an adjunct to astronaut "boots on the ground." The approach is to develop and propose controlled experiments, which build upon previous field studies and which will assess the effects of different latencies (0 to 500 msec), task complexity, and alternate forms of feedback to the operator. These experiments serve as an example of a pathfinder for NASA's roadmap of missions to Mars with low-latency telerobotic exploration as a precursor to astronaut's landing on the surface to conduct geological tasks.

ISS in Support↗
NASA NTRS2015

Low-Latency Teleoperations for Human Exploration and Evolvable Mars Campaign

NASA has been analyzing a number of mission concepts and activities that involve low-latency telerobotic (LLT) operations. One mission concept that will be covered in this presentation is Crew-Assisted Sample Return which involves the crew acquiring samples (1) that have already been delivered to space, and or acquiring samples via LLT from orbit to a planetary surface and then launching the samples to space to be captured in space and then returned to the earth with the crew. Both versions of have key roles for low-latency teleoperations. More broadly, the NASA Evolvable Mars Campaign is exploring a number of other activities that involve LLT, such as: (a) human asteroid missions, (b) PhobosDeimos missions, (c) Mars human landing site reconnaissance and site preparation, and (d) Mars sample handling and analysis. Many of these activities could be conducted from Mars orbit and also with the crew on the Mars surface remotely operating assets elsewhere on the surface, e.g. for exploring Mars special regions and or teleoperating a sample analysis laboratory both of which may help address planetary protection concerns. The operational and technology implications of low-latency teleoperations will be explored, including discussion of relevant items in the NASA Technology Roadmap and also how previously deployed robotic assets from any source could subsequently be used by astronauts via LLT.

crew autonomy↗
NASA NTRS2015

Message Latency Characterization of a Distributed Live, Virtual, Constructive Simulation Environment

A distributed test environment incorporating Live, Virtual, Constructive, (LVC) concepts was developed to execute standalone and integrated simulations and flight-tests that support unmanned aircraft research for NASA’s Unmanned Aircraft Systems (UAS) in the National Airspace System (NAS) Project. The LVC components form the core infrastructure that supports simulation of UAS operations by integrating live and virtual aircraft in a realistic air traffic environment. This LVC infrastructure enables efficient testing by leveraging the use of existing distributed assets. The LVC concepts used for the UAS in the NAS project include live aircraft, flight simulators, and virtual air traffic control assets operating at facilities distributed across multiple NASA Centers. With a distributed network, however, there is a concern that message latency could impact the realism of a simulation and its data. The latencies associated with sending data among these distributed facilities were, therefore, measured to ensure that they fall within acceptable parameters. Several live and virtual test assets were integrated into the LVC infrastructure including NASA Armstrong’s Ikhana MQ-9 unmanned aircraft, NASA Glenn’s S3-B manned aircraft, and the B747 flight simulator at NASA Ames. Average latencies from 100 to 150 milliseconds were observed between the LVC System running at NASA Ames and each of the participating NASA Centers under a light-to-moderate (fifty aircraft) traffic sample.

Murphy, James R.↗
NASA NTRS2019

EDOS Initiatives to Decrease Latency of NRT Data for LANCE

NASA's EOS Data and Operations System (EDOS) is the primary supplier of NRT (near real-time) data to the NASA NRT user community known as the Land, Atmosphere NRT Capability for EOS (LANCE). EDOS provides NRT data for various instruments on the EOS missions Terra, Aqua, Aura, as well as for the NOAA missions Suomi NPP and NOAA-20. This poster describes an overview of the EDOS multi-mission system with emphasis on the NRT products distributed for LANCE elements: AIRS, MISR, MLS, MODIS, MOPITT, OMPS, OMI and VIIRS. Remote EDOS high-rate data capture systems are deployed at NASA ground stations which provide data-driven capture of high-rate science for EOS missions. The remote EDOS components transfer the science data via high-rate WANs to the centralized EDOS Level-zero processing systems located at Goddard Space Flight Center. EDOS produces session-based data sets especially for LANCE NRT use from a single ground station contact; this data is sent to dual LANCE destinations as part of the standard redundancy requirement for LANCE elements. EDOS has implemented various latency improvements with the ultimate goal to have EDOS processing of NRT data keep up with the spacecraft data downlink. EDOS enhancements have included implementation of priority-based QoS, expanded network architecture to include open networks, and use of a delay-tolerant protocol. EDOS has streamlined its systems and infrastructure to minimize latency for NRT data delivery for LANCE. EDOS begins to transfer the NRT data to the LANCE elements within minutes of the end of the contact session with an average packet latency from instrument observation to Level 0 product delivery to each LANCE element of just over one hour.

Michael, Karen↗
NASA NTRSDate not supplied

Impact of Latency and Reliability on Separation Assurance with Remotely Piloted Aircraft in Terminal Operations

Remotely Piloted Aircraft (RPA) for cargo operations in the national airspace system will impact safety due to, among other factors, the latency and reliability of command & control, and of communication. This paper investigates the safety impact with increasing mix of RPA amidst manned traffic in a generic arrival pattern with three merging flows. Latency was modelled as the response time between air traffic control's determination of a resolution and the RPAs' initiation of the maneuver. Reliability was modelled as a message drop probability. The experiment was repeated with two different aircraft types having different performance characteristics as representatives of RPA for conducting automated cargo operations. Overall response time above thirty seconds and message drop probability over twenty percent caused losses of separation. Specific results depended on the RPA aircraft type. The detailed impacts of latency and reliability with increasing mix of RPA traffic are provided. Applications of the approach for further studies at increasing levels of automation are also discussed.

Vishwanath Bulusu↗
NASA NTRS2022

Impact of Latency and Reliability on Separation Assurance with Remotely Piloted Aircraft in Terminal Operations

Remotely Piloted Aircraft (RPA) for cargo operations in the national airspace system will impact safety due to, among other factors, the latency and reliability of command & control, and of communication. This paper investigates the safety impact with increasing mix of RPA amidst manned traffic in a generic arrival pattern with three merging flows. Latency was modelled as the response time between air traffic control's determination of a resolution and the RPAs' initiation of the maneuver. Reliability was modelled as a message drop probability. The experiment was repeated with two different aircraft types having different performance characteristics as representatives of RPA for conducting automated cargo operations. Overall response time above thirty seconds and message drop probability over twenty percent caused losses of separation. Specific results depended on the RPA aircraft type. The detailed impacts of latency and reliability with increasing mix of RPA traffic are provided. Applications of the approach for further studies at increasing levels of automation are also discussed.

Vishwanath Bulusu↗
DOE OSTI2024

Architectural Implications of Neural Network Inference for High Data-Rate, Low-Latency Scientific Applications

With more scientific fields relying on neural networks (NNs) to process data incoming at extreme throughputs and latencies, it is crucial to develop NNs with all their parameters stored on-chip. In many of these applications, there is not enough time to go off-chip and retrieve weights. Even more so, off-chip memory such as DRAM does not have the bandwidth required to process these NNs as fast as the data is being produced (e.g., every 25 ns). As such, these extreme latency and bandwidth requirements have architectural implications for the hardware intended to run these NNs: 1) all NN parameters must fit on-chip, and 2) codesigning custom/reconfigurable logic is often required to meet these latency and bandwidth constraints. In our work, we show that many scientific NN applications must run fully on chip, in the extreme case requiring a custom chip to meet such stringent constraints.

Weng, Olivia↗
DOE OSTI2023

Ultra-low latency recurrent neural network inference on FPGAs for physics applications with hls4ml

Abstract Recurrent neural networks have been shown to be effective architectures for many tasks in high energy physics, and thus have been widely adopted. Their use in low-latency environments has, however, been limited as a result of the difficulties of implementing recurrent architectures on field-programmable gate arrays (FPGAs). In this paper we present an implementation of two types of recurrent neural network layers—long short-term memory and gated recurrent unit—within the hls4ml framework. We demonstrate that our implementation is capable of producing effective designs for both small and large models, and can be customized to meet specific design requirements for inference latencies and FPGA resources. We show the performance and synthesized designs for multiple neural networks, many of which are trained specifically for jet identification tasks at the CERN Large Hadron Collider.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

MLL1 is regulated by KSHV LANA and is important for virus latency

Abstract Mixed lineage leukemia 1 (MLL1) is a histone methyltransferase. Kaposi's sarcoma-associated herpesvirus (KSHV) is a leading cause of malignancy in AIDS. KSHV latently infects tumor cells and its genome is decorated with epigenetic marks. Here, we show that KSHV latency-associated nuclear antigen (LANA) recruits MLL1 to viral DNA where it establishes H3K4me3 modifications at the extensive KSHV terminal repeat elements during primary infection. LANA interacts with MLL1 complex members, including WDR5, integrates into the MLL1 complex, and regulates MLL1 activity. We describe the 1.5-Å crystal structure of N-terminal LANA peptide complexed with MLL1 complex member WDR5, which reveals a potential regulatory mechanism. Disruption of MLL1 expression rendered KSHV latency establishment highly deficient. This deficiency was rescued by MLL1 but not by catalytically inactive MLL1. Therefore, MLL1 is LANA regulable and exerts a central role in virus infection. These results suggest broad potential for MLL1 regulation, including by non-host factors.

Tan, Min↗
DOE OSTI2023

Latency and Bandwidth Microbenchmarks of US Department of Energy Systems in the June 2023 Top 500 List

As a rule, Top 500 class supercomputers are extensively benchmarked as part of their acceptance testing process. However, barring publicly posted LINPACK / HPCG results, most benchmark results are often inaccessible outside the hosting institution. Moreover, these higher level benchmarks do not provide easy answers to common questions such as “What is the realizable memory bandwidth?” or “What is the launch latency on the accelerator?” To partially address these issues, we executed selected single-node micro-benchmarks — focused on latencies and memory bandwidth — on every US Department of Energy system above rank 150 of the June 2023 Top 500 list, excepting NERSC’s Cori and ORNL’s Frontier TDS (now decommissioned or repurposed). We hope to provide an easy “first stop” reference for users of current Top 500 systems and inspire users and administrators of other Top 500 systems to similarly compile and make available benchmark results for their systems.

Siefert, Christopher M.↗
DOE OSTI2021

Ps and Qs: Quantization-Aware Pruning for Efficient Low Latency Neural Network Inference

Efficient machine learning implementations optimized for inference in hardware have wide-ranging benefits, depending on the application, from lower inference latency to higher data throughput and reduced energy consumption. Two popular techniques for reducing computation in neural networks are pruning, removing insignificant synapses, and quantization, reducing the precision of the calculations. In this work, we explore the interplay between pruning and quantization during the training of neural networks for ultra low latency applications targeting high energy physics use cases. Techniques developed for this study have potential applications across many other domains. We study various configurations of pruning during quantization-aware training, which we term quantization-aware pruning, and the effect of techniques like regularization, batch normalization, and different pruning schemes on performance, computational complexity, and information content metrics. We find that quantization-aware pruning yields more computationally efficient models than either pruning or quantization alone for our task. Further, quantization-aware pruning typically performs similar to or better in terms of computational efficiency compared to other neural architecture search techniques like Bayesian optimization. Surprisingly, while networks with different training configurations can have similar performance for the benchmark application, the information content in the network can vary significantly, affecting its generalizability.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
NASA NTRS1976

Long-latency evoked potentials to irrelevant, deviant stimuli

Occasional shifts of loudness in a repetitive train of clicks elicited a late-positive wave (P3a) in nonattending subjects which peaked at a mean latency of 258 msec and had a frontocentral scalp distribution; P3a was typically preceded by an 'N2' component at 196 msec. The P3a wave was distinguishable from the longer-latency (378 msec) parietocentrally distributed 'P3b' wave that was evoked by the same stimulus in an actively attending subject, thus confirming the findings of Squires et al. (1975). Infrequently presented single sounds did not produce large or consistent N2-P3a components; the critical condition for the generation of an N2-P3a wave seemed to be that the infrequent sounds represent a deviation (intensity increment or decrement) from a repetitive background. Furthermore, increasing the repetition rate of the background clicks drastically reduced N1-P2 amplitude but had little effect on the amplitude of N2-P3a. This suggests that N2-P3a is not simply a delayed N1-P2 'vertex potential', but rather reflects the operation of a 'mismatch' detector, which registers deviations from an ongoing auditory background.

Snyder, E.↗
NASA NTRS1983

New results in fault latency modelling

Studies carried out by McGough and Swern (1981, 1983) are summarized. In these studies, an avionics processor was simulated and a series of fault injection experiments was carried out to determine the degree of fault latency in a redundant flight control system that employed comparison monitoring as the exclusive means of failure detection. A determination was also made of the fault coverage of a typical self-test program. The summary presented stresses that a self-test program should be designed to capitalize on the hardware mechanization of the processor. If this is not done, subtests tend to repeatedly exercise the same hardware components while neglecting to exercise a substantial proportion of the remainder. It is also pointed out that fault latency is relatively independent of both the length and instruction mix of a program. A significant difference is found in fault coverage assessed using pin-level and gate-level fault models.

Mcgough, J. G.↗
NASA NTRS1984

Far-field brainstem responses evoked by vestibular and auditory stimuli exhibit increases in interpeak latency as brain temperature is decreased

The effect of decreasing of brain temperature on the brainstem auditory evoked response (BAER) in rats was investigated. Voltage pulses, applied to a piezoelectric crystal attached to the skull, were used to evoke stimuli in the auditory system by means of bone-conducted vibrations. The responses were recorded at 37 C and 34 C brain temperatures. The peaks of the BAER recorded at 34 C were delayed in comparison with the peaks from the 37 C wave, and the later peaks were more delayed than the earlier peaks. These results indicate that an increase in the interpeak latency occurs as the brain temperature is decreased. Preliminary experiments, in which responses to brief angular acceleration were used to measure the brainstem vestibular evoked response (BVER), have also indicated increases in the interpeak latency in response to the lowering of brain temperature.

Hoffman, L. F.↗
NASA NTRS1986

Effect of macular ablation on frequency and latency of motion-induced emesis in the squirrel monkey

Three previously motion-emetic sensitive squirrel monkeys were rendered refractory to a standard motion-emetic regimen by a two-stage utriculosacculectomy procedure which preserved the cristae ampullares of semicircular canals. Three nonoperated control squirrel monkeys tested on the same motion-emetic regimen time schedule as the operated animals remained motion-emetic sensitive with regard to incidence, frequency, and latency of motion-induced emetic responses. Following a sham surgical procedure (stapedectomy) performed on two of the latter animals and one additional new animal, the emetic incidence decreased from 100 to 89 percent, but the frequency and latency were not altered significantly.

Brizzee, Kenneth R.↗
NASA NTRS1991

Robo-line storage: Low latency, high capacity storage systems over geographically distributed networks

Rapid advances in high performance computing are making possible more complete and accurate computer-based modeling of complex physical phenomena, such as weather front interactions, dynamics of chemical reactions, numerical aerodynamic analysis of airframes, and ocean-land-atmosphere interactions. Many of these 'grand challenge' applications are as demanding of the underlying storage system, in terms of their capacity and bandwidth requirements, as they are on the computational power of the processor. A global view of the Earth's ocean chlorophyll and land vegetation requires over 2 terabytes of raw satellite image data. In this paper, we describe our planned research program in high capacity, high bandwidth storage systems. The project has four overall goals. First, we will examine new methods for high capacity storage systems, made possible by low cost, small form factor magnetic and optical tape systems. Second, access to the storage system will be low latency and high bandwidth. To achieve this, we must interleave data transfer at all levels of the storage system, including devices, controllers, servers, and communications links. Latency will be reduced by extensive caching throughout the storage hierarchy. Third, we will provide effective management of a storage hierarchy, extending the techniques already developed for the Log Structured File System. Finally, we will construct a protototype high capacity file server, suitable for use on the National Research and Education Network (NREN). Such research must be a Cornerstone of any coherent program in high performance computing and communications.

Katz, Randy H.↗
NASA NTRS1999

Short latency compound action potentials from mammalian gravity receptor organs

Gravity receptor function was characterized in four mammalian species using far-field vestibular evoked potentials (VsEPs). VsEPs are compound action potentials of the vestibular nerve and central relays that are elicited by linear acceleration ramps applied to the cranium. Rats, mice, guinea pigs, and gerbils were studied. In all species, response onset occurred within 1.5 ms of the stimulus onset. Responses persisted during intense (116 dBSPL) wide-band (50 to 50 inverted question mark omitted inverted question mark000 Hz) forward masking, whereas auditory responses to intense clicks (112 dBpeSPL) were eliminated under the same conditions. VsEPs remained after cochlear extirpation but were eliminated following bilateral labyrinthectomy. Responses included a series of positive and negative peaks that occurred within 8 ms of stimulus onset (range of means at +6 dBre: 1.0 g/ms: P1=908 to 1062 micros, N1=1342 to 1475 micros, P2=1632 to 1952 micros, N2=2038 to 2387 micros). Mean response amplitudes at +6 dBre: 1.0 g/ms ranged from 0.14 to 0.99 microV. VsEP input/output functions revealed latency slopes that varied across peaks and species ranging from -19 to -51 micros/dB. Amplitude-intensity slopes also varied ranging from 0.04 to 0.08 microV/dB for rats and mice. Latency values were comparable to those of birds although amplitudes were substantially smaller in mammals. VsEP threshold values were considerably higher in mammals compared to birds and ranged from -8.1 to -10.5 dBre 1.0 g/ms across species. These results support the hypothesis that mammalian gravity receptors are less sensitive to dynamic stimuli than are those of birds.

Non-NASA Center↗