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 451 records · Page 25
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

In-depth analysis on parallel processing patterns for high-performance Dataframes

The Data Science domain has expanded monumentally in both research and industry communities during the past decade, predominantly owing to the Big Data revolution. Artificial Intelligence (AI) and Machine Learning (ML) are bringing more complexities to data engineering applications, which are now integrated into data processing pipelines to process terabytes of data. Typically, a significant amount of time is spent on data preprocessing in these pipelines, and hence improving its efficiency directly impacts the overall pipeline performance. The community has recently embraced the concept of Dataframes as the de-facto data structure for data representation and manipulation. However, the most widely used serial Dataframes today (R, pandas) experience performance limitations while working on even moderately large data sets. We believe that there is plenty of room for improvement by taking a look at this problem from a high-performance computing point of view. In a prior publication, we presented a set of parallel processing patterns for distributed dataframe operators and the reference runtime implementation, Cylon. In this paper, we are expanding on the initial concept by introducing a cost model for evaluating the said patterns. Furthermore, we evaluate the performance of Cylon on the ORNL Summit supercomputer.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Distributed Tomographic Reconstruction with Quantization

Conventional tomographic reconstruction typically depends on centralized servers for both data storage and computation, leading to concerns about memory limitations and data privacy. Distributed reconstruction algorithms mitigate these issues by partitioning data across multiple nodes, reducing server load and enhancing privacy. However, these algorithms often encounter challenges related to memory constraints and communication overhead between nodes. In this paper, we introduce a decentralized Alternating Directions Method of Multipliers (ADMM) with configurable quantization. By distributing local objectives across nodes, our approach is highly scalable and can efficiently reconstruct images while adapting to available resources. To overcome communication bottlenecks, we propose two quantization techniques based on K-means clustering and JPEG compression. Numerical experiments with benchmark images illustrate the tradeoffs between communication efficiency, memory use, and reconstruction accuracy.

Miao, Runxuan↗
DOE OSTI2025

Fourier-based three-dimensional multistage transformer for aberration correction in multicellular specimens

High-resolution tissue imaging is often compromised by sample-induced optical aberrations that degrade resolution and contrast. Although wavefront sensor-based adaptive optics (AO) can measure these aberrations, such hardware solutions are typically complex, expensive to implement and slow when serially mapping spatially varying aberrations across large fields of view. Here we introduce AOViFT (adaptive optical vision Fourier transformer)—a machine learning-based aberration sensing framework built around a three-dimensional multistage vision transformer that operates on Fourier domain embeddings. AOViFT infers aberrations and restores diffraction-limited performance in puncta-labeled specimens with substantially reduced computational cost, training time and memory footprint compared to conventional architectures or real-space networks. We validated AOViFT on live gene-edited zebrafish embryos, demonstrating its ability to correct spatially varying aberrations using either a deformable mirror or postacquisition deconvolution. By eliminating the need for the guide star and wavefront sensing hardware and simplifying the experimental workflow, AOViFT lowers technical barriers for high-resolution volumetric microscopy across diverse biological samples.

Alshaabi, Thayer [Howard Hughes Medical Institute,↗
DOE OSTI2022

Ultraefficient resistance switching between charge ordered phases in 1 T -TaS 2 with a single picosecond electrical pulse

Progress in high-performance computing demands significant advances in memory technology. Among novel memory technologies that promise efficient device operation on a sub-ns timescale, resistance switching between charge ordered phases of 1T-TaS 2 has shown to be potentially useful for development of high-speed, energy efficient nonvolatile memory devices. Measurement of the electrical operation of such devices in the picosecond regime is technically challenging and hitherto still largely unexplored. Here, we use an optoelectronic “laboratory-on-a-chip” experiment for measurement of ultrafast memory switching, enabling accurate measurement of electrical switching parameters with 100 fs temporal resolution. Photoexcitation and electro-optic sampling on a (Cd,Mn)Te substrate are used to generate and, subsequently, measure electrical pulse propagation with intra-band excitation and sub-gap probing, respectively. We demonstrate high contrast nonvolatile resistance switching from high to low resistance states of a 1T-TaS 2 device using single sub-2ps electrical pulses. Using detailed modeling, we find that the switching energy density per unit area is exceptionally small, E A =9.4 fJ/μm 2 . Finally, the speed and energy efficiency of an electronic “write” process place the 1T-TaS 2 devices into a category of their own among new generation nonvolatile memory devices.

42 ENGINEERING↗
DOE OSTI2025

Stabilization of magnetic bubbles in [Ni/Co]n multilayers on an oxygen-reconstructed Nb(110) surface via an ultra-thin Cu interlayer

Magnetic thin films hosting topological spin textures, such as magnetic skyrmions, hold high potential for breakthroughs in the field of spintronics, due to good scalability and energy efficiency. Novel computational architectures such as memory-in-logic devices rely on material platforms able to host those topological spin textures. Furthermore, recently proposed designs of novel quantum information technologies are based on heterostructures where topological spin textures are in direct proximity to a superconducting layer. Here, we demonstrate the stabilization of out-of-plane magnetic bubbles in highly ordered [Ni/Co]n multilayers on a Nb(110) single crystal. This is achieved without the need for the removal of the well-known Nb(110)-oxide surface reconstruction, due to the introduction of an ultra-thin Cu interlayer in between the Nb substrate and the magnetic multilayer. The Cu interlayer generates a well-ordered hexagonal surface, which is key for the epitaxial growth of the [Ni/Co]n multilayers hosting the desired out-of-plane anisotropy. The magnetic ground state of the prepared material stacks is directly imaged via spin-polarized low-energy electron microscopy, revealing the presence of magnetic bubble domains with lateral sizes as small as 450 nm.

Dibajeh, Ahmad↗
DOE OSTI2022

Electronegative metal dopants improve switching variability in Al 2 ⁢O 3 resistive switching devices

Resistive random-access memories are promising for nonvolatile memory and brain-inspired computing applications. High variability and low yield of these devices are key drawbacks hindering reliable training of physical neural networks. In this paper, we show that doping an oxide electrolyte, Al 2 ⁢O 3 , with electronegative metals makes resistive switching significantly more reproducible, surpassing the reproducibility requirements for obtaining reliable hardware neuromorphic circuits. Based on density functional theory calculations, the underlying mechanism is hypothesized to be the ease of creating oxygen vacancies in the vicinity of electronegative dopants due to the capture of the associated electrons by dopant midgap states and the weakening of Al-O bonds. These oxygen vacancies and vacancy clusters also bind significantly to the dopant, thereby serving as preferential sites and building blocks in the formation of conducting paths. Throughout this work, we validate this theory experimentally by implanting different dopants over a range of electronegativities in devices made of multiple alternating layers of Al 2 ⁢O 3 and WN and find superior repeatability and yield with highly electronegative metals, Au, Pt, and Pd. These devices also exhibit a gradual SET transition, enabling multibit switching that is desirable for analog computing.

36 MATERIALS SCIENCE↗
DOE OSTI2020

Resource Management for Processing Wide Area Data Streams on Supercomputers

Modern scientific instruments generate enormous amount of data. Typically, the data collected from the instruments are stored in one or more files that are then moved to a distant supercomputer for processing. The final results are sent back to the user. In order to make effective use of the time on expensive instruments, experimenters want to process the data as they are generated. They want to stream the data from instruments' memory directly to a supercomputer's memory for analysis. Since the compute nodes in a supercomputer are not connected directly to the wide area network, the data streams need to be passed through intermediate gateway nodes. As opposed to the best effort file transfers, data streaming applications require resources at a specific time for a specific period. In this paper, we present a system model for enabling data streaming through gateway nodes and an algorithm to efficiently allocate gateway node resources along with compute nodes. We evaluate the algorithm using real-world traces on the Chameleon Cloud. The results show that our system can schedule compute and gateway resources efficiently for streaming analysis.

data streaming↗
DOE OSTI2022

Scalable Deep Learning-Based Microarchitecture Simulation on GPUs

Cycle-accurate microarchitecture simulators are essential tools for designers to architect, estimate, optimize, and manufacture new processors that meet specific design expectations. However, conventional simulators based on discrete-event methods often require an exceedingly long time-to-solution for the simulation of applications and architectures at full complexity and scale. Given the excitement around wielding the machine learning (ML) hammer to tackle various architecture problems, there have been attempts to employ ML to perform architecture simulations, such as Ithemal and SimNet. However, the direct application of existing ML approaches to architecture simulation may be even slower due to overwhelming memory traffic and stringent sequential computation logic. This work proposes the first graphics processing unit (GPU)-based microarchitecture simulator that fully unleashes the potential of GPUs to accelerate state-of-the-art ML-based simulators. First, considering the application traces are loaded from central processing unit (CPU) to GPU for simulation, we introduce various designs to reduce the data movement cost between CPUs and GPUs. Second, we propose a parallel simulation paradigm that partitions the application trace into sub-traces to simulate them in parallel with rigorous error analysis and effective error correction mechanisms. Combined, this scalable GPU-based simulator outperforms by orders of magnitude the traditional CPU-based simulators and the state-of-the-art ML-based simulators, i.e., SimNet and Ithemal.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Effect of image resolution on automated classification of chest X-rays

Deep learning (DL) models have received much attention lately for their ability to achieve expert-level performance on the accurate automated analysis of chest X-rays (CXRs). Recently available public CXR datasets include high resolution images, but state-of-the-art models are trained on reduced size images due to limitations on graphics processing unit memory and training time. As computing hardware continues to advance, it has become feasible to train deep convolutional neural networks on high-resolution images without sacrificing detail by downscaling. This study examines the effect of increased resolution on CXR classification performance. We used the publicly available MIMIC-CXR-JPG dataset, comprising 377,110 high resolution CXR images for this study. We applied image downscaling from native resolution to 2048 × 2048 pixels, 1024 × 1024 pixels, 512 × 512 pixels, and 256 × 256 pixels and then we used the DenseNet121 and EfficientNet-B4 DL models to evaluate clinical task performance using these four downscaled image resolutions. We find that while some clinical findings are more reliably labeled using high resolutions, many other findings are actually labeled better using downscaled inputs. We qualitatively verify that tasks requiring a large receptive field are better suited to downscaled low resolution input images, by inspecting effective receptive fields and class activation maps of trained models. Lastly, we show that stacking an ensemble across resolutions outperforms each individual learner at all input resolutions while providing interpretable scale weights, indicating that diverse information is extracted across resolutions.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

A Hybrid Finite-Volume, Discontinuous Galerkin Discretization for the Radiative Transport Equation

In this report we propose a hybrid spatial discretization for the radiative transport equation that combines a second-order discontinuous Galerkin (DG) method and a second-order finite-volume (FV) method. The strategy relies on a simple operator splitting that has been used previously to combine different angular discretizations. Unlike standard FV methods with upwind fluxes, the hybrid approach is able to accurately simulate problems in scattering dominated regimes. However, it requires less memory and yields a faster computational time than a uniform DG discretization. In addition, the underlying splitting allows naturally for hybridization in both space and angle. Numerical results are given to demonstrate the efficiency of the hybrid approach in the context of discrete ordinate angular discretizations and Cartesian spatial grids.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

On-the-fly response function generation method for composite coarse mesh

The hybrid stochastic deterministic transport code COMET, based on the incident response expansion theory, is used to model reactor cores with high fidelity and formidable computational speed. COMET models a reactor core using a library of incident flux response expansion coefficients that are pre computed for all the unique lattice cells (e.g., fuel assemblies, reflector blocks, etc.) in the core. In order to further improve its computational efficiency in pre-calculating the response library a new response function generation method is developed to compute the response functions for the composite coarse meshes made of a smaller set of unique lattices on the fly within the COMET's deterministic transport core sweep. The efficiency is achieved by eliminating a number of unique lattices that can be made up from the reduced set of unique meshes on the fly. The numerical process consists of the following steps. First, the boundary condition on composite coarse mesh boundaries is projected onto the expansion basis to compute the incident flux moments on external surfaces of all the basic (reduced set of unique) coarse meshes. Secondly, the deterministic sweeping solver in COMET is used to converge on the outgoing/incoming flux expansion moments crossing interfaces between the basic coarse meshes. Thirdly, the response functions for the composite coarse meshes are constructed as a superposition on the fly. The new response function generation method was tested on 88 composite coarse meshes consisting of CANDU fuel bundles and moderator blocks. It was found that response functions generated by the new method agree very well with those generated by direct Monte Carlo calculations. The average and maximum relative differences in the surface-to-surface response coefficients computed by the two methods are 0.10% and 0.20%, respectively. Similarly, the average and maximum relative differences in the response fission densities are 0.13% and 0.43%, respectively. These discrepancies are within one standard deviation of the stochastic uncertainties. The new method is five times faster than the original direct Monte Carlo method. The size of the response function library for the new method is five times smaller than that for the original method, leading to significantly less requirement for the computer hard drive space and memory. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2020

A view into FleCSI [Slides]

FleCSI is largely based around sharing of memory spaces. FleCSI facilitates parallel execution and sharing of memory across memory spaces. Execute() transforms sender-side data structures into receiver-side data structures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Accelerating geostatistical modeling using geostatistics-informed machine Learning

Ordinary Kriging (OK) is a popular geostatistical algorithm for spatial interpolation and estimation. The computational complexity of OK changes quadratically and cubically for memory and speed, respectively, given the number of data. Therefore, it is computationally intensive and also challenging to process a large set of data, especially in three-dimensional (3D) cases. This paper develops a geostatistics-informed machine learning (GIML) model to improve the efficiency of OK by reducing the number of points required to be estimated using OK. Specifically, only a very few of the unknown points are estimated by OK to get the weights and estimations, which are used as the training dataset. Moreover, the governing equations of OK are used to guide our proposed machine learning to better reproduce the spatial distributions. Our results show that the proposed GIML can reduce the computational time of OK by at least one order of magnitude. The effectiveness of the GIML is evaluated and compared using a 2D case. Furthermore, we demonstrate its efficiency and robustness by considering a different number of training samples on various 3D simulation grids.

58 GEOSCIENCES↗
DOE OSTI2021

Oineus v1.0

A library for multi-threaded computation of persistence diagrams. The algorithm for computing persistence diagrams in a lock-free manner was published in the 'Towards Lock-free Persistent Homology' (D. Morozov, A. Nigmetov, Brief Announcement: SPAA 2020); it scales better than the only other shared-memory parallel implementation of persistent homology computation PHAT. Library includes python bindings to compute persistence diagrams and their vectorizations for lower-star filtrations on grid data. It is intended to be used by scientists working on Topological Data Analysis and its applications in different areas.

Nigmetov, Arnur↗
DOE OSTI2020

Logic in Memory Emulator

Logic in Memory Emulator (LiME) is a hardware/software tool specially designed for memory system evaluation and experiment. Emerging memories display a wide range of bandwidths, latencies, and capacities, making it challenging for the computer architect to navigate the design space of potential memory configurations, and for the application developer to assess performance implications of using such memories. With the LiME framework, architectural ideas can be prototyped in great detail yet with sufficient performance to support realistic evaluation on long running applications. LiME consists of two fundamental components: 1) the hardware and OS infrastructure for the emulator, and 2) a suite of benchmark applications to assist in characterizing the performance of current and future computer architectures. Some of the applications have been collected from other open source projects. Uses: Logging, replay and analysis of an application's memory behavior Evaluate impact of emerging memory technology on application performance. Emulate complex memory interactions in whole applications orders of magnitude faster than software simulation. Emulate acceleration hardware co-located with the memory subsystem. Features: Capture and log external memory accesses to a separate off-chip memory device without affecting application execution. Memory traces include the address, length, timestamp, and optionally the data for each transaction. Captured trace data can be saved to an SD card for off-line analysis. Configure a wide range of memory latencies in sub-nanosecond increments that encompass highbandwidth and storage class memories. Specify regions of interest (ROI) in applications to reduce the amount of trace data captured for analysis. Currently supports execution on Xilinx Zynq SoC which integrates an ARM processor with FPGA logic on a single device. Applications can be run under Linux or in bare metal mode on the ARM cores.

Jain, AbhishekK↗
DOE OSTI2023

Spatially Distributed Ramp Reversal Memory in VO 2

Ramp‐reversal memory has recently been discovered in several insulator‐to‐metal transition materials where a non‐volatile resistance change can be set by repeatedly driving the material partway through the transition. This study uses optical microscopy to track the location and internal structure of accumulated memory as a thin film of VO 2 is temperature cycled through multiple training subloops. These measurements reveal that the gain of insulator phase fraction between consecutive subloops occurs primarily through front propagation at the insulator‐metal boundaries. By analyzing transition temperature maps, it is found, surprisingly, that the memory is also stored deep inside both insulating and metallic clusters throughout the entire sample, making the metal‐insulator coexistence landscape more rugged. This non‐volatile memory is reset after heating the sample to higher temperatures, as expected. Diffusion of point defects is proposed to account for the observed memory writing and subsequent erasing over the entire sample surface. By spatially mapping the location and character of non‐volatile memory encoding in VO 2 , this study results enable the targeting of specific local regions in the film where the full insulator‐to‐metal resistivity change can be harnessed in order to maximize the working range of memory elements for conventional and neuromorphic computing applications.

36 MATERIALS SCIENCE↗
DOE OSTI2021

Butterfly Factorization Via Randomized Matrix-Vector Multiplications

This paper presents an adaptive randomized algorithm for computing the butterfly factorization of an m × n matrix with m ≈ n provided that both the matrix and its transpose can be rapidly applied to arbitrary vectors. The resulting factorization is composed of O(log n) sparse factors, each containing O(n) nonzero entries. The factorization can be attained using O(n 3/2 log n) computation and O(n log n) memory resources. Furthermore, the proposed algorithm can be implemented in parallel and can apply to matrices with strong or weak admissibility conditions arising from surface integral equation solvers as well as multi-frontal-based finite-difference, finite-element, or finite-volume solvers. A distributed-memory parallel implementation of the algorithm demonstrates excellent scaling behavior.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Impacts of noise and structure on quantum information encoded in a quantum memory

As larger, higher-quality quantum devices are built and demonstrated in quantum information applications, such as quantum computation and quantum communication, the need for high-quality quantum memories to store quantum states becomes ever more pressing. Future quantum devices likely will use a variety of physical hardware, some being used primarily for processing of quantum information and others for storage. Here we study the correlation of the structure of quantum information with physical noise models of various possible quantum memory implementations. Through numerical simulation of different noise models and approximate analytical formulas applied to a variety of interesting quantum states, we provide comparisons between quantum hardware with different structure, including both qubit- and qudit-based quantum memories. Our findings point to simple, experimentally relevant formulas for the relative lifetimes of quantum information in different quantum memories and have relevance to the design of hybrid quantum devices.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗