Search NASA⌕ Search

SEARCH · Search NASA

Results for “asynchronous methods”

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 181 records · Page 10
All sourcesNASA NTRSDOE OSTI
NASA NTRS2011

NASA Tech Briefs, August 2011

Topics covered include: Miniature, Variable-Speed Control Moment Gyroscope; NBL Pistol Grip Tool for Underwater Training of Astronauts; HEXPANDO Expanding Head for Fastener-Retention Hexagonal Wrench; Diagonal-Axes Stage for Pointing an Optical Communications Transceiver; Improvements in Speed and Functionality of a 670-GHz Imaging Radar; IONAC-Lite; Large Ka-Band Slot Array for Digital Beam-Forming Applications; Development of a 150-GHz MMIC Module Prototype for Large-Scale CMB Radiation; Coupling Between Waveguide-Fed Slot Arrays; PCB-Based Break-Out Box; Multiple-Beam Detection of Fast Transient Radio Sources; Router Agent Technology for Policy-Based Network Management; Remote Asynchronous Message Service Gateway; Automatic Tie Pointer for In-Situ Pointing Correction; Jitter Correction; MSLICE Sequencing; EOS MLS Level 2 Data Processing Software Version 3; DspaceOgre 3D Graphics Visualization Tool; Metallization for Yb14MnSb11-Based Thermoelectric Materials; Solvent/Non-Solvent Sintering To Make Microsphere Scaffolds; Enhanced Fuel-Optimal Trajectory-Generation Algorithm for Planetary Pinpoint Landing; Self-Cleaning Coatings and Materials for Decontaminating Field-Deployable Land and Water-Based Optical Systems; Separation of Single-Walled Carbon Nanotubes with DEP-FFF; Li Anode Technology for Improved Performance; Post-Fragmentation Whole Genome Amplification-Based Method; Microwave Tissue Soldering for Immediate Wound Closure; Principles, Techniques, and Applications of Tissue Microfluidics; Robotic Scaffolds for Tissue Engineering and Organ Growth; Stress-Driven Selection of Novel Phenotypes; Method for Accurately Calibrating a Spectrometer Using Broadband Light; Catalytic Microtube Rocket Igniter; Stage Cylindrical Immersive Display; Vacuum Camera Cooler; Atomic Oxygen Fluence Monitor; Thermal Management Tools for Propulsion System Trade Studies and Analysis; Introduction to Physical Intelligence; Technique for Solving Electrically Small to Large Structures for Broadband Applications; Accelerated Adaptive MGS Phase Retrieval; Large Eddy Simulation Study for Fluid Disintegration and Mixing; Tropospheric Correction for InSAR Using Interpolated ECMWF Data and GPS Zenith Total Delay; Technique for Calculating Solution Derivatives With Respect to Geometry Parameters in a CFD Code; Acute Radiation Risk and BRYNTRN Organ Dose Projection Graphical User Interface; Probabilistic Path Planning of Montgolfier Balloons in Strong, Uncertain Wind Fields; Flight Simulation of ARES in the Mars Environment; Low-Outgassing Photogrammetry Targets for Use in Outer Space; Planning the FUSE Mission Using the SOVA Algorithm; Monitoring Spacecraft Telemetry Via Optical or RF Link; and Robust Thermal Control of Propulsion Lines for Space Missions.

Source record↗
DOE OSTI2022

The L-CAPE Project at FNAL

The controls system at FNAL records data asynchronously from several thousand Linac devices at their respective cadences, ranging from 15Hz down to once per minute. In case of downtimes, current operations are mostly reactive, investigating the cause of an outage and labeling it after the fact. However, as one of the most upstream systems at the FNAL accelerator complex, the Linac’s foreknowledge of an impending downtime as well as its duration could prompt downstream systems to go into standby, potentially leading to energy savings. The goals of the Linac Condition Anomaly Prediction of Emergence (L-CAPE) project that started in late 2020 are (1) to apply data-analytic methods to improve the information that is available to operators in the control room, and (2) to use machine learning to automate the labeling of outage types as they occur and discover patterns in the data that could lead to the prediction of outages. We present an overview of the challenges in dealing with time-series data from 2000+ devices, our approach to developing an ML-based automated outage labeling system, and the status of augmenting operations by identifying the most likely devices predicting an outage.

43 PARTICLE ACCELERATORS↗
NASA NTRSDate not supplied

A Zero-order Hold Approach for Fractional-delay Interpolation in Auralization

During the signal processing chain of an auralization simulating the propagation of a sound from a moving source to a stationary receiver, it is often necessary to interpolate between the samples of the source signal in order to arrive at uniformly-spaced samples at the receiver. In some cases, this interpolation is done in the receiver time frame – where the “input” samples of the source have become irregularly spaced due to time dilation effects. Canonical band-limited interpolation methods (i.e., sinc and sinc-derived approaches) cannot be applied in this case as they rely on having a uniformly-spaced input. The use of geometric interpolation methods that can handle irregularly-spaced input may not be grounded in signal processing principles and may produce unwanted artifacts and noise. This presentation outlines the possibility of embedding an irregularly-spaced zero-order hold signal within a highly over-sampled uniformly-spaced signal, and then processing down to the desired sampling rate through successive decimations. Initial distortion and noise characteristics of the approach are shown for some basic propagation geometries. The possible benefits of using such an approach in an auralization scheme with a time-varying Doppler shift are discussed including: the prevention of aliasing, processing time advantages, and the possibility for asynchronous processing.

Auralization↗
DOE OSTI2021

Proton-detected solid-state NMR spectroscopy of spin-1/2 nuclei with large chemical shift anisotropy

Constant-time (CT) dipolar heteronuclear multiple quantum coherence (D-HMQC) has previously been demonstrated as a method for proton detection of high-resolution wideline NMR spectra of spin-1/2 nuclei with large chemical shift anisotropy (CSA). However, 1 H transverse relaxation and t 1 -noise often reduce the sensitivity of D-HMQC experiments, preventing the theoretical gains in sensitivity provided by 1 H detection from being realized. In this paper, we demonstrate a series of improved pulse sequences for 1 H detection of spin-1/2 nuclei under fast MAS, with 195 Pt SSNMR experiments on cisplatin as an example. First, a t 1 -incrementation protocol for D-HMQC dubbed Arbitrary Indirect Dwell (AID) is demonstrated. AID allows the use of arbitrary, rotor asynchronous t 1 -increments, but removes the constant time period from CT D-HMQC, resulting in improved sensitivity by reducing transverse relaxation losses. Next, we show that short high-power adiabatic pulses (SHAPs), which efficiently invert broad MAS sideband manifolds, can be effectively incorporated into 1 H detected symmetry-based resonance echo double resonance (S-REDOR) and t 1 -noise eliminated (TONE) D-HMQC experiments. The S-REDOR experiments with SHAPs provide approximately double the dipolar dephasing, as compared to experiments with rectangular inversion pulses. We lastly show that sensitivity and resolution can be further enhanced with the use of swept excitation pulses as well as adiabatic magic angle turning (aMAT).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2024

Efficient distributed continual learning for steering experiments in real-time

Deep learning has emerged as a powerful method for extracting valuable information from large volumes of data. However, when new training data arrives continuously (i.e., is not fully available from the beginning), incremental training suffers from catastrophic forgetting (i.e., new patterns are reinforced at the expense of previously acquired knowledge). Training from scratch each time new training data becomes available would result in extremely long training times and massive data accumulation. Rehearsal-based continual learning has shown promise for addressing the catastrophic forgetting challenge, but research to date has not addressed performance and scalability. To fill this gap, we propose an approach based on a distributed rehearsal buffer that efficiently complements data-parallel training on multiple GPUs to achieve high accuracy, short runtime, and scalability. It leverages a set of buffers (local to each GPU) and uses several asynchronous techniques for updating these local buffers in an embarrassingly parallel fashion, all while handling the communication overheads necessary to augment input minibatches using unbiased, global sampling. We further propose a generalization of rehearsal buffers to support both classification and generative learning tasks, as well as more advanced rehearsal strategies (notably Dark Experience Replay, leveraging knowledge distillation). We illustrate this approach with a real-life HPC streaming application from the domain of ptychographic image reconstruction. Furthermore, we run extensive experiments on up to 128 GPUs of the ThetaGPU supercomputer to compare our approach with baselines representative of training-from-scratch (the upper bound in terms of accuracy) and incremental training (the lower bound). Results show that rehearsal-based continual learning achieves a top-5 validation accuracy close to the upper bound, while simultaneously exhibiting a runtime close to the lower bound.

Asynchronous data management↗
NASA NTRS2012

Extreme Tele-Echocardiography: Methodology for Remote Guidance of In-Flight Echocardiography Aboard the International Space Station

Methods: In the year before launch of an ISS mission, potential astronaut echocardiographic operators participate in 5 sessions to train for echo acquisitions that occur roughly monthly during the mission, including one exercise echocardiogram. The focus of training is familiarity with the study protocol and remote guidance procedures. On-orbit, real-time guidance of in-flight acquisitions is provided by a sonographer in the Telescience Center of Mission Control. Physician investigators with remote access are able to relay comments on image quality to the sonographer. Live video feed is relayed from the ISS to the ground via the Tracking and Data Relay Satellite System with a 2- second transmission delay. The expert sonographer uses these images, along with twoway audio, to provide instructions and feedback. Images are stored in non-compressed DICOM format for asynchronous relay to the ground for subsequent off-line analysis. Results: Since June, 2009, a total of 27 resting echocardiograms and 5 exercise studies have been performed during flight. Average acquisition time has been 45 minutes, reflecting 26,000 km of ISS travel per study. Image quality has been adequate in all studies, and remote guidance has proven imperative for fine-tuning imaging and prioritizing views when communication outages limit the study duration. Typical resting studies have included 27 video loops and 30 still-frame images requiring 750 MB of storage. Conclusions: Despite limited crew training, remote guidance allows research-quality echocardiography to be performed by non-experts aboard the ISS. Analysis is underway and additional subjects are being recruited to define the impact of microgravity on cardiac structure and systolic and diastolic function.

Martin, David S.↗
NASA NTRS2011

Extreme Tele-Echocardiography: Methodology for Remote Guidance of In-flight Echocardiography Aboard the International Space Station

Echocardiography is ideally suited for cardiovascular imaging in remote environments, but the expertise to perform it is often lacking. In 2001, an ATL HDI5000 was delivered to the International Space Station (ISS). The instrument is currently being used in a study to investigate the impact of long-term microgravity on cardiovascular function. The purpose of this report is to describe the methodology for remote guidance of echocardiography in space. Methods: In the year before launch of an ISS mission, potential astronaut echocardiographic operators participate in 5 sessions to train for echo acquisitions that occur roughly monthly during the mission, including one exercise echocardiogram. The focus of training is familiarity with the study protocol and remote guidance procedures. On-orbit, real-time guidance of in-flight acquisitions is provided by a sonographer in the Telescience Center of Mission Control. Physician investigators with remote access are able to relay comments on image optimization to the sonographer. Live video feed is relayed from the ISS to the ground via the Tracking and Data Relay Satellite System with a 2 second transmission delay. The expert sonographer uses these images along with two-way audio to provide instructions and feedback. Images are stored in non-compressed DICOM format for asynchronous relay to the ground for subsequent off-line analysis. Results: Since June, 2009, a total of 19 resting echocardiograms and 4 exercise studies have been performed in-flight. Average acquisition time has been 45 minutes, reflecting 26,000 km of ISS travel per study. Image quality has been adequate in all studies, but remote guidance has proven imperative for fine-tuning imaging and prioritizing views when communication outages limit the study duration. Typical resting studies have included 12 video loops and 21 still-frame images requiring 750 MB of storage. Conclusions: Despite limited crew training, remote guidance allows research-quality echocardiography to be performed by non-experts aboard the ISS. Analysis is underway and additional subjects are being recruited to define the impact of microgravity on cardiac structure and systolic and diastolic function.

Martin, David↗
DOE OSTI2021

Three practical workflow schedulers for easy maximum parallelism

Runtime scheduling and workflow systems are an increasingly popular algorithmic component in HPC because they allow full system utilization with relaxed synchronization requirements. There are so many special-purpose tools for task scheduling, one might wonder why more are needed. Use cases seen on the Summit supercomputer needed better integration with MPI and greater flexibility in job launch configurations. Preparation, execution, and analysis of computational chemistry simulations at the scale of tens of thousands of processors revealed three distinct workflow patterns. A separate job scheduler was implemented for each one using extremely simple and robust designs: file-based, task-list based, and bulk-synchronous. Comparing to existing methods shows unique benefits of this work, including simplicity of design, suitability for HPC centers, short startup time, and well-understood per-task overhead. All three new tools have been shown to scale to full utilization of Summit, and have been made publicly available with tests and documentation. This work presents a complete characterization of the minimum effective task granularity for efficient scheduler usage scenarios. Here, these schedulers have the same bottlenecks, and hence similar task granularities as those reported for existing tools following comparable paradigms.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Asynchronous Navier-Stokes Solver on 3D Unstructured Grids for the Exascale Era

This project has developed multiple fluid dynamics solvers for complex 3D flows using fully asynchronous distributed-memory task-parallel algorithms on top of the Charm++ runtime system. The algorithms solve the Euler or Navier-Stokes equations of compressible flows using unstructured tetrahedron meshes with optional solution-adaptive mesh-, and polynomial-order refinement. We have demonstrated excellent strong scaling up 50K compute cores and the benefits of Charm++’s automatic load balancing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Accelerating Application Bulk Synchronous Writes in HPC Environments

High-bandwidth storage tiers are becoming more common for their capability to absorb high-rate, bursty I/Os. Notably, the designs of these fast storage tiers differ from system to system. The variation of these layers and non-uniform methods of access can pose chal- lenges for applications seeking to run at multiple HPC facilities. Therefore, in this work, we present Spectral, a rapid-output ab- straction library to accelerate application, bulk-synchronous writes on HPC systems. We design Spectral to enable applications to use high-bandwidth storage, such as node-local storage and dis- tributed, write-caches (e.g., burst buffers) transparently without requiring modifications to the application or file system source code. The key idea is to allow applications to spend most of the time performing productive work and to not require any source code changes for maximum portability on different HPC archi- tectures. Spectral internally re-routes write-only files through available, high-performance I/O resources before ultimately mi- grating them to the shared global parallel file system. For instance, on Summit, Spectral transparently places application outputs on node-local storage and then utilizes asynchronous migration to the center-wide GPFS file system. We evaluate Spectral on the Summit HPC system (1024 nodes) using the IOR benchmark and real scientific applications. Spectral shows linear performance scaling, improving application write performance by over an order of magnitude when compared to GPFS.

Khan, Awais↗
NASA NTRS2006

The Effectiveness of TAG or Guard-Gates in SET Suppression Using Delay and Dual-Rail Configurations at 0.35 microns

Design options for decreasing the susceptibility of integrated circuits to Single Event Upset (SEU) fall into two categories: (1) increasing the critical charge to cause an upset at a particular node, and (2) employing redundancy to mask or correct errors. With decreasing device sizes on an Integrated Circuit (IC), the amount of charge required to represent a logic state has steadily reduced. Critical charge methods such as increasing drive strength or increasing the time required to change state as in capacitive or resistive hardening or delay based approaches extract a steadily increasing penalty as a percentage of device resources and performance. Dual redundancy is commonly assumed only to provide error detection with Triple Modular Redundancy (TMR) required for correction, but less well known methods employ dual redundancy to achieve full error correction by voting two inputs with a prior state to resolve ambiguity. This requires special circuits such as the Whitaker latch [1], or the guard-gate [2] which some of us have called a Transition AND Gate (TAG) [3]. A 2-input guard gate is shown in Figure 1. It is similar to a Muller Completion Element [4] and relies on capacitance at node "out" to retain the prior state when inputs disagree, while eliminating any output buffer which would be susceptible to radiation strikes. This paper experimentally compares delay based and dual rail flip-flop designs wherein both types of circuits employ guard-gates to optimize layout and performance, and draws conclusions about design criteria and suitability of each option. In both cases a design goal is protection against Single Event Transients (SET) in combinational logic as well as SEU in the storage elements. For the delay based design, it is also a goal to allow asynchronous clear or preset inputs on the storage elements, which are often not available in radiation tolerant designs.

Shuler, Robert L.↗
DOE OSTI2024

Radiation-Induced Noise Resilience of Neuromorphic Architectures

Neuromorphic event-based networks use asynchronous time-dependent information to extract features from input data that can allow for edge-based distributed applications such as object recognition. The noise resilience properties of such networks, especially in the context of space applications, are yet to be explored. In this paper, we use the hierarchy of time surfaces (HOTS) algorithm, which is one of the neuromorphic algorithms, to understand the least and most resilient modules in a neuromorphic network. The HOTS algorithm relies on the computing of time surfaces that maps the temporal delays between neighboring pixels into normalized features that involve many computations that are also found in other neuromorphic networks such as exponential decays, distance computations, etcetera. We implemented HOTS on a Digilent PYNQ board with a Xilinx Zynq 7020 system on a chip, and we subjected the boards running the HOTS network inference to neutron radiation at the Los Alamos Neutron Science Center. Furthermore, we used simulation models from our previous similar experiments on the event-based sensor to create a neutron induced noise model to quantify the effect of this noise on the overall performance of the network. This experiment provides the preliminary measurements of the reliability of the HOTS algorithm and proposes methods to create a more reliable HOTS architecture in future spacecraft missions.

Engineering↗
NASA NTRS1997

Left ventricular diastolic filling with an implantable ventricular assist device: beat to beat variability with overall improvement

OBJECTIVES: We studied the effects of left ventricular (LV) unloading by an implantable ventricular assist device on LV diastolic filling. BACKGROUND: Although many investigators have reported reliable systemic and peripheral circulatory support with implantable LV assist devices, little is known about their effect on cardiac performance. METHODS: Peak velocities of early diastolic filling, late diastolic filling, late to early filling ratio, deceleration time of early filling, diastolic filling period and atrial filling fraction were measured by intraoperative transesophageal Doppler echocardiography before and after insertion of an LV assist device in eight patients. A numerical model was developed to simulate this situation. RESULTS: Before device insertion, all patients showed either a restrictive or a monophasic transmitral flow pattern. After device insertion, transmitral flow showed rapid beat to beat variation in each patient, from abnormal relaxation to restrictive patterns. However, when the average values obtained from 10 consecutive beats were considered, overall filling was significantly normalized from baseline, with early filling velocity falling from 87 +/- 31 to 64 +/- 26 cm/s (p < 0.01) and late filling velocity rising from 8 +/- 11 to 32 +/- 23 cm/s (p < 0.05), resulting in an increase in the late to early filling ratio from 0.13 +/- 0.18 to 0.59 +/- 0.38 (p < 0.01) and a rise in the atrial filling fraction from 8 +/- 10% to 26 +/- 17% (p < 0.01). The deceleration time (from 112 +/- 40 to 160 +/- 44 ms, p < 0.05) and the filling period corrected by the RR interval (from 39 +/- 8% to 54 +/- 10%, p < 0.005) were also significantly prolonged. In the computer model, asynchronous LV assistance produced significant beat to beat variation in filling indexes, but overall a normalization of deceleration time as well as other variables. CONCLUSIONS: With LV assistance, transmitral flow showed rapidly varying patterns beat by beat in each patient, but overall diastolic filling tended to normalize with an increase of atrial contribution to the filling. Because of the variable nature of the transmitral flow pattern with the assist device, the timing of the device cycle must be considered when inferring diastolic function from transmitral flow pattern.

Non-NASA Center↗
DOE OSTI2025

Myna: Connecting powder bed fusion build data to simulation tools for digital twin applications

Additive manufacturing (AM), as a digital process, can generate a detailed digital thread linking a part’s design and manufacturing to its operational performance. As AM systems advance, an increasing amount of process data is stored in manufacturing databases. In principle, this data can be utilized by simulation-based digital twin approaches, such as real-time process control and asynchronous post-processing guidance. However, few tools currently exist for systematically integrating digital thread data with computational tools. Here, in this study, we propose a software package, called Myna, for connecting data from powder bed fusion processes to simulation tools. The utility of such a platform is demonstrated using build data from the Oak Ridge National Laboratory Manufacturing Demonstration Facility “Peregrine v2023-10” public dataset to automatically configure and run 54 semi-analytical 3DThesis melt pool simulations, 78 numerical Additive FOAM melt pool simulations, and 3 ExaCA microstructure simulations. The simulated, spatially registered microstructures are then compared directly with electron backscatter diffraction characterization of the corresponding as-built part locations. The resulting simulated microstructure showed variation as a function of process parameters, particularly stripe width; however, the experimental data had little variation between the microstructure texture and grain size resulting from different processing conditions. Analysis of the discrepancies suggest that it is possible a two-phase ferritic-austenitic solidification model is needed to accurately predict grain size and texture for certain stainless steel 316L feedstock compositions under powder bed fusion conditions, providing direction for future research. As illustrated here, due to the number and complexity of the simulations involved in AM process-structure–property predictions, automated methods to connect process data and simulations will remain necessary tools for testing hypotheses and implementing digital twin applications.

Knapp, Gerald L. [Oak Ridge National Laboratory (O↗
DOE OSTI2022

Computer Science Research Needs for Parallel Discrete Event Simulation (PDES)

Historically, scientific computing efforts have demonstrated the clear need for, and effective use of, supercomputing with traditional time-stepped simulations. Nevertheless, there are several areas in the mission spaces of the U.S. Department of Energy and other agencies waiting to tap advanced computing research using a different, discrete event style of modeling, simulation, and analysis. These span a wide spectrum of applications including energy grid resilience, urban planning and policy, transportation science, building technologies, emergency response and planning, environmental impact analysis, computational epidemiology, Internet communications, cyber security, and cyber-physical systems, to name only a few. Even within traditional scientific applications, the role of discrete event modes of execution is increasing in the form of new event-based mathematical solvers such as quantized state integration methods and discrete-continuous hybrid system solvers. Co-design of advanced supercomputing hardware systems is another area that exploits discrete event simulation at its core for effective analyses. Complex systems, entity behaviors and interconnections play a significant role in all these applications, which are mapped to large-scale models with discrete event formulations. To make advancements in all the aforementioned scientific areas, many technical aspects need to be more thoroughly studied and deeply understood in parallel discrete event simulation (PDES). The unique dynamics inherent in a discrete event modeling approach, by their very nature, intersect and influence the entire stack of the computing system, including (a) the unique nature of the instruction sets exercised in PDES workloads without a predominance of high-precision floating point operations, (b) virtual time-constrained multi-threaded execution of many logical processes per processor, (c) extremely variable and difficult to predict network traffic characteristics, (d) interfaces and inter-dependencies with machine learning and artificial intelligence codes at higher software layers, and (e) highly challenging load balancing needs, especially in effectively accounting for accelerated/extremely heterogeneous computing in current and future high-performance computing systems. Efficient and accurate parallel execution of PDES workloads is also dominated by challenges in dealing with their asynchronous concurrency fundamentally present at the model level. Conservative synchronization, optimistic/speculative synchronization, and their hybrid schemes open new questions in fundamental computer science with respect to reversibility of computation and prediction (lookahead) of behaviors inherent within model codes. On the implementation front, there are relatively few scalable, general-purpose parallel discrete event simulators in the world, and even fewer have been studied on emerging hardware platforms. To enable scientific advances using PDES, the research needs in computer science must also be pursued and met in the intersection of the algorithmic and hardware-aware aspects of scalable PDES engines. This report is aimed at capturing a computer science-oriented view of this important area of research in PDES, presenting a sample of important applications with their inherent discrete event technology elements. Needs are outlined in core areas of parallel discrete event research as well as cross-cutting directions in computer science research that positively impact scientific advancements across several important application areas. A selection of priority research opportunities in advanced computing for PDES is identified to serve as reference for key research topics and their order of importance for scientific advancements.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Robustness of Deep Learning Classification to Adversarial Input on GPUs: Asynchronous Parallel Accumulation Is a Source of Vulnerability

The ability of machine learning (ML) classification models to resist small, targeted input perturbations—known as adversarial attacks—is a key measure of their safety and reliability. We show that floating-point non associativity (FPNA) coupled with asynchronous parallel programming on GPUs is sufficient to result in misclassification, without any perturbation to the input. Additionally, we show that this misclassification is particularly significant for inputs close to the decision boundary and that standard adversarial robustness results may be overestimated up to 4.6 when not considering machine-level details. We first study a linear classifier, before focusing on standard Graph Neural Network (GNN) architectures and datasets used in robustness assessments. We develop a novel black-box attack using Bayesian optimization to discover external workloads that can change the instruction scheduling which bias the output of reductions on GPUs and reliably lead to misclassification. Motivated by these results, we present a new learnable permutation (LP) gradient-based approach to learning floating-point operation orderings that lead to misclassifications. The LP approach provides a worst-case estimate in a computationally efficient manner, avoiding the need to run identical experiments tens of thousands of times over a potentially large set of possible GPU states or architectures. Finally, using instrumentation-based testing, we investigate parallel reduction ordering across different GPU architectures under external background workloads, when utilizing multi-GPU virtualization, and when applying power capping. Our results demonstrate that parallel reduction ordering varies significantly across architectures under the first two conditions, substantially increasing the search space required to fully test the effects of this parallel scheduler-based vulnerability. These results and the methods developed here can help to include machine-level considerations into adversarial robustness assessments, which can make a difference in safety and mission critical applications.

Shanmugavelu, Sanjif [Maxeler Technologies, a Groq↗
DOE OSTI2024

Characteristics of Vertical Ground Motions and Their Effect on the Seismic Response of Bridges in the Near-Field: A State-of-the-Art Review

Despite the evidence from past earthquakes and several numerical investigations demonstrating the detrimental impact of vertical ground motions (VGMs) on the integrity of bridge structures, incorporating their effects into seismic assessment and design procedures has traditionally been given limited consideration. Current codes utilize rather simplistic approaches to account for the concurrent effects of vertical and horizontal motions in structural performance evaluations, potentially leading to unconservative estimates of structural demands. This paper reviews the main features of VGMs and their effect on the seismic response of bridges. The methods and empirical models available to estimate vertical motions for design purposes are discussed, and research gaps and related research needs are identified. Finally, the emerging role of physics-based ground-motion simulations, as well as their limitations, in supporting future research and informing the development of simplified design procedures is examined. The main areas of interest for future research are identified in the need to carry out systematic sensitivity studies to gain insight into the main earthquake parameters that influence key VGMs features, understand the influence of soil nonlinearities on VGMs amplitude and frequency content, inform the development of empirical models that cover a range of site conditions and source-to-site distances where current models are poorly constrained, generate arrays of motions to update coherency models to properly inform the analysis of distributed infrastructure, investigate the impulsive character of VGMs, and assess the approximations made in estimating VGMs with 1D site response analyses. Furthermore, specific focus is laid on large-magnitude earthquakes in the near-field.

Asynchronous ground motion↗
DOE OSTI2022

Integrating a ponderomotive guiding center algorithm into a quasi-static particle-in-cell code based on azimuthal mode decomposition

High fidelity modeling of plasma based acceleration (PBA) requires the use of three dimensional, fully nonlinear, and kinetic descriptions based on the particle-in-cell (PIC) method. In PBA an intense particle beam or laser (driver) propagates through a tenuous plasma whereby it excites a plasma wave wake. Three-dimensional PIC algorithms based on the quasi-static approximation (QSA) have been successfully applied to efficiently model the interaction between relativistic charged particle beams and plasma. In a QSA PIC algorithm, the plasma response to a charged particle beam or laser driver is calculated based on forces from the driver and self-consistent forces from the QSA form of Maxwell's equations. These fields are then used to advance the charged particle beam or laser forward by a large time step. Since the time step is not limited by the regular Courant-Friedrichs-Lewy (CFL) condition that constrains a standard 3D fully electromagnetic PIC code, a 3D QSA PIC code can achieve orders of magnitude speedup in performance. Recently, a new hybrid QSA PIC algorithm that combines another speedup technique known as an azimuthal Fourier decomposition has been proposed and implemented. This hybrid algorithm decomposes the electromagnetic fields, charge and current density into azimuthal harmonics and only the Fourier coefficients need to be updated, which can reduce the algorithmic complexity of a 3D code to that of a 2D code. Modeling the laser-plasma interaction in a full 3D electromagnetic PIC algorithm is very computationally expensive due the enormous disparity of physical scales to be resolved. In the QSA the laser is modeled using the ponderomotive guiding center (PGC) approach. We describe how to implement a PGC algorithm compatible for the QSA PIC algorithms based on the azimuthal mode expansion. Here this algorithm permits time steps orders of magnitude larger than the cell size and it can be asynchronously parallelized. Details on how this is implemented into the QSA PIC code that utilizes an azimuthal mode expansion, QPAD, are also described. Benchmarks and comparisons between a fully 3D explicit PIC code (OSIRIS), as well as a few examples related to laser wakefield acceleration, are presented.

97 MATHEMATICS AND COMPUTING↗