Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “compute in memory”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Electrically Controlled All-Antiferromagnetic Tunnel Junctions on Silicon with Large Room-Temperature Magnetoresistance

Antiferromagnetic (AFM) materials are a pathway to spintronic memory and computing devices with unprecedented speed, energy efficiency, and bit density. Realizing this potential requires AFM devices with simultaneous electrical writing and reading of information, which are also compatible with established silicon-based manufacturing. Recent experiments have shown tunneling magnetoresistance (TMR) readout in epitaxial AFM tunnel junctions. However, these TMR structures are not grown using a silicon-compatible deposition process, and controlling their AFM order required external magnetic fields. Here are shown three-terminal AFM tunnel junctions based on the noncollinear antiferromagnet PtMn 3 , sputter-deposited on silicon. The devices simultaneously exhibit electrical switching using electric currents, and electrical readout by a large room-temperature TMR effect. First-principles calculations explain the TMR in terms of the momentum-resolved spin-dependent tunneling conduction in tunnel junctions with noncollinear AFM electrodes.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Scalable Al 2 O 3 –TiO 2 Conductive Oxide Interfaces as Defect Reservoirs for Resistive Switching Devices

Abstract Resistive switching devices herald a transformative technology for memory and computation, offering considerable advantages in performance and energy efficiency. Here, a simple and scalable material system of conductive oxide interfaces is employed, and their unique properties are leveraged for a new type of resistive switching device. An Al 2 O 3 –TiO 2 ‐based valence‐change resistive switching device, where the conductive oxide interface serves both as the bottom electrode and as a reservoir of defects for switching, is demonstrated. The amorphous–polycrystalline Al 2 O 3 –TiO 2 conductive interface is obtained following the technological path of simplifying the fabrication of the 2D electron gases (2DEGs), making them scalable for practical mass integration. Physical analysis of the device chemistry and microstructure with comprehensive electrical analysis of its switching behavior and performance is combined. The origin of the resistive switching is pinpointed to the conductive oxide interface, which serves both as the bottom electrode and as a reservoir of oxygen vacancies. The latter plays a key role in valence‐change resistive switching devices. The new device, based on scalable and complementary metal–oxide–semiconductor (CMOS)‐technology‐compatible fabrication processes, opens new design spaces toward increased tunability and simplification of the device selection challenge.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Design and implementation of I/O performance prediction scheme on HPC systems through large-scale log analysis

Abstract Large-scale high performance computing (HPC) systems typically consist of many thousands of CPUs and storage units used by hundreds to thousands of users simultaneously. Applications from large numbers of users have diverse characteristics, such as varying computation, communication, memory, and I/O intensity. A good understanding of the performance characteristics of each user application is important for job scheduling and resource provisioning. Among these performance characteristics, I/O performance is becoming increasingly important as data sizes rapidly increase and large-scale applications, such as simulation and model training, are widely adopted. However, predicting I/O performance is difficult because I/O systems are shared among all users and involve many layers of software and hardware stack, including the application, network interconnect, operating system, file system, and storage devices. Furthermore, updates to these layers and changes in system management policy can significantly alter the I/O behavior of applications and the entire system. To improve the prediction of the I/O performance on HPC systems, we propose integrating information from several different system logs and developing a regression-based approach to predict the I/O performance. Our proposed scheme can dynamically select the most relevant features from the log entries using various feature selection algorithms and scoring functions, and can automatically select the regression algorithm with the best accuracy for the prediction task. The evaluation results show that our proposed scheme can predict the write performance with up to 90% prediction accuracy and the read performance with up to 99% prediction accuracy using the real logs from the Cori supercomputer system at NERSC.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Acceleration of Graph Neural Network-Based Prediction Models in Chemistry via Co-Design Optimization on Intelligence Processing Units

Atomic structure prediction and associated property calculations are the bedrock of chemical physics. Since high-fidelity ab initio modeling techniques for computing the structure and properties can be prohibitively expensive, this motivates the development of machine-learning (ML) models that make these predictions more efficiently. Training graph neural networks over large atomistic databases introduces unique computational challenges such as the need to process millions of small graphs with variable size and support communication patterns that are distinct from learning over large graphs such as social networks. We demonstrate a novel hardware-software co-design approach to scale up the training of atomistic graph neural networks (GNN) for structure and property prediction. First, to eliminate redundant computation and memory associated with alternative padding techniques and to improve throughput via minimizing communication, we formulate the effective coalescing of the batches of variable-size atomistic graphs as the bin packing problem and introduce a hardware-agnostic algorithm to pack these batches. In addition, we propose hardware-specific optimizations including a planner and vectorization for the gather-scatter operations targeted for Graphcore’s Intelligence Processing Unit (IPU), as well as model-specific optimizations such as merged communication collectives and optimized softplus. Putting these all together, we demonstrate the effectiveness of the proposed co-design approach by providing an implementation of a well-established atomistic GNN on the Graphcore IPUs. We evaluate the training performance on multiple atomistic graph databases with varying degrees of graph counts, sizes and sparsity. Here, we demonstrate that such a co-design approach can reduce the training time of atomistic GNNs and can improve the performance by up to 1.5× compared to the baseline implementation of the model on the IPUs. Additionally, we compare our IPU implementation with a Nvidia GPU-based implementation and show that our atomistic GNN implementation on the IPUs can run 1.8× faster on average compared to the execution time on the GPUs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

Electrode and Microstructure Dependence of Oxygen Diffusion in Ferroelectric Hafnium Zirconium Oxide Thin Films

Hafnia-based ferroelectrics hold promise to reduce energy demand for computing by enabling compute-in-memory and as non-volatile memories. The ferroelectric phase in this material system is, in part, stabilized by oxygen vacancies. While oxygen vacancies may be a necessity for phase stability, they limit device endurance through diffusion and accumulation into conducting channels. Herein, it is shown that oxygen diffusion is spatially variable within individual grains of ferroelectric hafnium zirconium oxide (HZO). Using 18 O tracers and finite difference modeling, it is shown that grain boundaries and regions near electrode interfaces allow for relatively rapid oxygen diffusion, with values as much as 10 4 larger than the grain cores. Further, the selection of electrode material affects the diffusion coefficients across all microstructural regions. HZO films in contact with TiN electrodes result in more oxygen-deficient HZO films and higher oxygen diffusion coefficients. Tungsten electrodes result in fewer vacancies and lower diffusion coefficients. Diffusion activation energy differences between the HZO with the two electrodes is reconciled by differing populations of charged and uncharged oxygen vacancies. This insight into the local vacancy populations and diffusion pathways provides a platform for designing hafnia-based films, deposition processes, and integration strategies to reduce vacancy gradients and improve performance.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Dynamic Symmetry Breaking in Chiral Magnetic Systems

The Dzyaloshinskii–Moriya interaction (DMI) in magnetic systems stabilizes spin textures with preferred chirality, applicable to next-generation memory and computing architectures. In perpendicularly magnetized heavy-metal/ferromagnet films, the interfacial DMI originating from structural inversion asymmetry and strong spin-orbit coupling favors chiral Néel-type domain walls (DWs) whose energetics and mobility remain at issue. Thus, a new effect is characterized in which domains expand unidirectionally in response to a combination of out-of-plane and in-plane magnetic fields, with the growth direction controlled by the in-plane field strength. These growth directionalities and symmetries with applied fields cannot be understood from static treatments alone. The authors theoretically demonstrate that perpendicular field torques stabilize steady-state magnetization profiles highly asymmetric in elastic energy, resulting in a dynamic symmetry breaking consistent with the experimental findings. This phenomenon sheds light on the mechanisms governing the dynamics of Néel-type DWs and expands the utility of field-driven DW motion to probe and control chiral DWs.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Synaptic Functionality and Neuromorphic Information Processing in Membrane Ion Channel Junctions

The human brain performs complex memory and computational tasks with high energy efficiency by regulating ion transport through membrane channels. These signaling mechanisms have been inspiring the development of nanofluidic memristors that emulate synaptic behavior. Here, in this study, we describe a membrane ion channel synapse (MICS), constructed from aqueous droplets linked by gramicidin A channels, that achieves neuromorphic functionality. MICS exhibits memristive ion transport with hysteretic current–voltage behavior arising from voltage-dependent channel formation and ion transport dynamics. MICS emulates a range of synaptic behaviors including associative learning. We further demonstrate its application in reservoir computing by performing handwritten digit classification and tic-tac-toe game and explore the system parameters that improve the computational performance. This droplet-based biomimetic synapse offers a potentially scalable and energy-efficient platform for next-generation neuromorphic computing systems.

Droplet interface bilayer↗
DOE OSTI2025

Improving vertical detail in simulated temperature and humidity data using machine learning

Atmospheric models used for weather forecasting and climate predictions discretise the atmosphere onto a vertical grid. There are however atmospheric phenomena that occur on scales smaller than the thickness of those model layers. The formation of low-level clouds due to temperature inversions is an example. This leads to atmospheric models underestimating, or even missing, these clouds and their radiative effects. Using radiosonde observations as training data, a machine learning model is used to improve the vertical detail of modelled profiles of temperature and specific humidity. In addition, a physics-informed machine learning model is developed and compared to the traditional approach; showing improvements in the cloud fraction profiles calculated from its predictions. The vertically enhanced profiles also improve the representation of layers of convective inhibition and anomalous refractivity gradients. This work facilitates targeted improvements to the representation of certain atmospheric processes without the burden of increased memory and computational cost from increasing vertical resolution throughout the whole model.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Cross Inference of Throughput Profiles Using Micro Kernel Network Method

Dedicated network connections are being increasingly deployed in cloud, centralized and edge computing and data infrastructures, whose throughput profiles are critical indicators of the underlying data transfer performance. Due to the cost and disruptions to physical infrastructures, network emulators, such as Mininet, are often used to generate measurements needed to estimate throughput profiles, typically expressed as a function of the connection round trip time. The profiles estimated using measurements from such emulated networks are usually inaccurate for high bandwidth and high latency connections, since they do not accurately reflect the critical network transport dynamics mainly due to computing and memory constraints of the host. We present a machine learning (ML) method to estimate the throughput profiles using emulation measurements to closely match the testbed and production network profiles. In particular, we propose a micro Kernel Network (mKN) that provides baseline throughput measurements on the host running Mininet emulations, which are used to learn a regression map that converts them to the corresponding testbed measurement estimates. Once initially learned, this map is applied to measurements from subsequent network emulations on the same host. We present experimental measurements to illustrate this approach, and derive generalization equations for the proposed mKN-ML method. Using a four-site scenario emulation, we show the effectiveness of this method in providing accurate concave throughput profiles from inaccurate convex or non-smooth ones indicated by Mininet emulation.

Rao, Nageswara↗
DOE OSTI2024

Tensor network representation of non-abelian gauge theory coupled to reduced staggered fermions

We show how to construct a tensor network representation of the path integral for reduced staggered fermions coupled to a non-abelian gauge field in two dimensions. The resulting formulation is both memory and computation efficient because reduced staggered fermions can be represented in terms of a minimal number of tensor indices while the gauge sector can be approximated using Gaussian quadrature with a truncation. Numerical results obtained using the Grassmann TRG algorithm are shown for the case of SU(2) lattice gauge theory and compared to Monte Carlo results.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2023

Efficient multi-scale representation of visual objects using a biologically plausible spike-latency code and winner-take-all inhibition

Deep neural networks have surpassed human performance in key visual challenges such as object recognition, but require a large amount of energy, computation, and memory. In contrast, spiking neural networks (SNNs) have the potential to improve both the efficiency and biological plausibility of object recognition systems. Here we present a SNN model that uses spike-latency coding and winner-take-all inhibition (WTA-I) to efficiently represent visual stimuli using multi-scale parallel processing. Mimicking neuronal response properties in early visual cortex, images were preprocessed with three different spatial frequency (SF) channels, before they were fed to a layer of spiking neurons whose synaptic weights were updated using spike-timing-dependent-plasticity. We investigate how the quality of the represented objects changes under different SF bands and WTA-I schemes. We demonstrate that a network of 200 spiking neurons tuned to three SFs can efficiently represent objects with as little as 15 spikes per neuron. Furthermore, studying how core object recognition may be implemented using biologically plausible learning rules in SNNs may not only further our understanding of the brain, but also lead to novel and efficient artificial vision systems.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Ensemble Kalman filter for data assimilation coupled with low-resolution computations techniques applied in fluid dynamics

This paper presents an innovative Reduced-order model (ROM) for merging experimental and simulation data using data assimilation (DA) to estimate the "True" state of a fluid dynamics system, leading to more accurate predictions. Our methodology introduces a novel approach by implementing the ensemble Kalman filter (EnKF) within a reduced-dimensional framework, grounded in a robust theoretical foundation and applied to fluid dynamics. To address the substantial computational demands of DA, the proposed ROM employs low-resolution (LR) techniques to drastically reduce computational costs. This innovative approach involves downsampling datasets for DA computations, followed by an advanced reconstruction technique based on low-cost singular value decomposition (lcSVD). The lcSVD method, a key innovation in this paper, has never been applied to DA before and offers a highly efficient way to enhance resolution with minimal computational resources. Our results demonstrate significant reductions in both computation time and RAM usage through these LR techniques without compromising the accuracy of the estimations. For instance, in a turbulent test case, for a data compression rate of 15.9, the LR approach can achieve a speed-up of 13.7 and a RAM compression of 90.9% while maintaining a low relative root mean square error (RRMSE) of 2.6%, compared to 0.8% in the high-resolution (HR) reference. Furthermore, we highlight the effectiveness of the EnKF in estimating and predicting the state of fluid flow systems based on limited observations and given low-fidelity numerical data. This paper highlights the potential of the proposed DA method in fluid dynamics applications, particularly for improving computational efficiency in CFD and related fields. Its ability to balance accuracy with low computational and memory costs makes it especially suitable for large-scale and real-time applications, such as environmental monitoring or engineering design. This method will be incorporated into ModelFLOWs-app.

Data Assimilation↗
DOE OSTI2022

A fast matrix-free approach to the high-order control volume finite element method with application to low-Mach flow

Here, a fast matrix-free formulation of the control volume finite element method is presented, requiring much less memory and computational work than previous efforts. The method is implemented and evaluated as a solver for low-Mach flow, including the evaluation of a preconditioning strategy for the pressure Poisson equation. The efficiency and scaling with polynomial order is evaluated on simple turbulent flows of interest, with appropriate solution quality metrics, and compared with a reference node-centered finite volume discretization. For a turbulent channel flow test, we show improvement in computational work for a given accuracy with the high-order scheme. The performance on a GPU accelerated platform is also investigated, with benefit shown for the matrix-free discretization.

42 ENGINEERING↗
DOE OSTI2020

Implicit fast sweeping method for hyperbolic systems of conservation laws

Implicit time-accurate methods are often used to integrate stiff problems where explicit schemes impose severe time step restrictions. This paper presents an efficient numerical framework based on the Fast Sweeping Method (FSM) for solving linear and nonlinear hyperbolic systems of conservation laws. The solution at each discrete location is computed by sweeping the numerical domain in several predetermined directions that follow the causality of the characteristic families. The use of a fractional step strategy eliminates the need for a solution selection criterion while one-sided stencils limit the number of sweeps to at most 2 d for d space dimensions. This work focuses on the first-order implicit upwind method since it constitutes the building block for high-order conservative schemes. For problems where the degree of stiffness evolves over time, implicit-explicit hybridization can be accomplished with the same algorithm by simply switching the stencil at each time level. As opposed to traditional implicit solvers, the sweeping method does not require a local time linearization of the fluxes thereby preserving the nonlinear stability properties of the original implicit scheme. It also avoids the large computational and memory requirements associated with solving large block-diagonal systems of equations. Here, a series of one- and two-dimensional test cases are presented for the inviscid Burgers' equation and the reactive Euler equations. The results indicate that the implicit FSM can allow a major reduction in the number of time steps even in the presence of discontinuous solution profiles.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2021

Targeted Writing and Deleting of Magnetic Skyrmions in Two-Terminal Nanowire Devices

Controllable writing and deleting of nanoscale magnetic skyrmions are key requirements for their use as information carriers for next-generation memory and computing technologies. While several schemes have been proposed, they require complex fabrication techniques or precisely tailored electrical inputs, which limits their long-term scalability. In this work, we demonstrate an alternative approach for writing and deleting skyrmions using conventional electrical pulses within a simple, two-terminal wire geometry. X-ray microscopy experiments and micromagnetic simulations establish the observed skyrmion creation and annihilation as arising from Joule heating and Oersted field effects of the current pulses, respectively. The unique characteristics of these writing and deleting schemes, such as spatial and temporal selectivity, together with the simplicity of the two-terminal device architecture, provide a flexible and scalable route to the viable applications of skyrmions.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2020

3D Nanomagnetism in Low Density Interconnected Nanowire Networks

Free-standing, interconnected metallic nanowire networks with densities as low as 40 mg/cm 3 have been achieved over centimeter-scale areas, using electrodeposition into polycarbonate membranes that have been ion-tracked at multiple angles. Networks of interconnected magnetic nanowires further provide an exciting platform to explore 3-dimensional nanomagnetism, where their structure, topology, and frustration may be used as additional degrees of freedom to tailor the materials properties. New magnetization reversal mechanisms in cobalt networks are captured by the first-order reversal curve method, which demonstrate the evolution from strong demagnetizing dipolar interactions to intersection-mediated domain wall pinning and propagation, and eventually to shape-anisotropy dominated magnetization reversal. These findings open up new possibilities for 3-dimensional integrated magnetic devices for memory, complex computation, and neuromorphics.

36 MATERIALS SCIENCE↗
DOE OSTI2023

Infrared Signatures for Phase Identification in Hafnium Oxide Thin Films

Phase identification in HfO 2 -based thin films is a prerequisite to understanding the mechanisms stabilizing the ferroelectric phase in these materials, which hold great promise in next-generation non-volatile memory and computing technology. While grazing-incidence X-ray diffraction is commonly employed for this purpose, it has difficulty unambiguously differentiating between the ferroelectric phase and other metastable phases that may exist due to similarities in the d-spacings, their low intensities, and the overlapping of reflections. Infrared signatures provide an alternative route. However, their use in phase identification remains limited because phase control has overwhelmingly been accomplished via substituents, thereby convoluting infrared signatures between the substituent and the phase changes they induce. Herein, we report the infrared optical responses of three undoped hafnium oxide films where annealing conditions have been used to create films consisting primarily of the ferroelectric polar orthorhombic Pca2 1 , antipolar orthorhombic Pbca, and monoclinic P2 1 /c phases, as was confirmed via transmission electron microscopy (TEM), UVvisible optical properties, and electrical property measurements. Vibrational signatures acquired from synchrotron nano-Fourier transform infrared spectroscopy (nano-FTIR) are shown to be capable of differentiating between the phases in a non-destructive, rapid, and nanoscale manner. The utility of nano-FTIR is illustrated for a film exhibiting an antiferroelectric polarization response. In this sample, it is proven that this behavior results from the Pbca phase rather than the often-cited tetragonal phase. Here, by demonstrating that IRspectroscopy can unambiguously distinguish phases in this material, this work establishes a tool needed to isolate the factors dictating ferroelectric phase stability in HfO 2 -based materials.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Néel domain walls with bistable chirality in a perpendicularly magnetized ferrimagnetic insulator

Field-free spin-orbit torque-driven domain wall motion in magnetic thin films with perpendicular magnetic anisotropy (PMA) requires the domain walls to have Néel character. Conventionally, Néel domain walls are stabilized by the Dzyaloshinskii-Moriya interaction (DMI) in ultrathin films. Here, in a europium iron garnet thin film with PMA and an additional uniaxial in-plane anisotropy, we demonstrate two bistable Néel domain wall states in the absence of DMI, and the capability to toggle the wall states with an in-plane field pulse and consequently their directions of motion under a current pulse. We present a phase diagram for the bistable Néel domain wall states as a function of in-plane field pulse width and amplitude. By fitting the experimental data to an analytical model of Néel wall reversal through the nucleation and propagation of Bloch lines, we extract the length of the initial reversed domain wall segment and Bloch line nucleation energy barrier. Current-driven motion of in-plane anisotropy stabilized Néel walls is qualitatively different from that of DMI-stabilized ones owing to the different symmetry of the effective fields that stabilize the Néel configuration. Furthermore, we present a proof of principle demonstration for 2-bit random number generation based on the stochastic reversal of domain wall chirality. These results provide critical insight into the topological energy barrier of Bloch lines and identify paths towards domain wall-based memory and computing devices.

36 MATERIALS SCIENCE↗