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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,063 records · Page 59

All sources NASA NTRS DOE OSTI

DOE OSTI2022

Natural switches in behaviour rapidly modulate hippocampal coding

Throughout their daily lives, animals and humans often switch between different behaviours. However, neuroscience research typically studies the brain while the animal is performing one behavioural task at a time, and little is known about how brain circuits represent switches between different behaviours. Here we tested this question using an ethological setting: two bats flew together in a long 135 m tunnel, and switched between navigation when flying alone (solo) and collision avoidance as they flew past each other (cross-over). Bats increased their echolocation click rate before each cross-over, indicating attention to the other bat. Hippocampal CA1 neurons represented the bat’s own position when flying alone (place coding). Notably, during cross-overs, neurons switched rapidly to jointly represent the interbat distance by self-position. This neuronal switch was very fast—as fast as 100 ms—which could be revealed owing to the very rapid natural behavioural switch. The neuronal switch correlated with the attention signal, as indexed by echolocation. Interestingly, the different place fields of the same neuron often exhibited very different tuning to interbat distance, creating a complex non-separable coding of position by distance. Theoretical analysis showed that this complex representation yields more efficient coding. Overall, our results suggest that during dynamic natural behaviour, hippocampal neurons can rapidly switch their core computation to represent the relevant behavioural variables, supporting behavioural flexibility.

60 APPLIED LIFE SCIENCES↗

DOE OSTI2024

Petahertz electronics

Petahertz, or lightwave, electronics uses tailored optical waveforms to control charge carriers in an electronic circuit at petahertz frequencies. This may enable faster processing than conventional pulsed electronics, which cannot be scaled beyond gigahertz frequencies. In recent years, petahertz-scale currents driven by optical fields have been measured in solid-state systems and nanoscale structures, with several proof-of-principle demonstrations of sub-optical-cycle current generation and optical-field-resolved waveform detection at the sub-femtosecond to few-femtosecond scale. Recent work has taken the first steps towards digital and quantum operation by exploring optical-field-driven logic and memory functionality. Here, in this Review, we discuss the progress towards sub-cycle field-driven current injection, highlighting key theoretical concepts, experimental milestones, and questions remaining as we push towards realizing petahertz electronics for ultrafast optical waveform analysis, digital logic, communications, and quantum computation.

Heide, Christian↗

DOE OSTI2020

Algorithms for Efficient Reproducible Floating Point Summation

We define “reproducibility” as getting bitwise identical results from multiple runs of the same program, perhaps with different hardware resources or other changes that should not affect the answer. Many users depend on reproducibility for debugging or correctness. However, dynamic scheduling of parallel computing resources, combined with nonassociative floating point addition, makes reproducibility challenging even for summation, or operations like the BLAS. We describe a “reproducible accumulator” data structure (the “binned number”) and associated algorithms to reproducibly sum binary floating point numbers, independent of summation order. We use a subset of the IEEE Floating Point Standard 754-2008 and bitwise operations on the standard representations in memory. Our approach requires only one read-only pass over the data, and one reduction in parallel, using a 6-word reproducible accumulator (more words can be used for higher accuracy), enabling standard tiling optimization techniques. Summing n words with a 6-word reproducible accumulator requires approximately 9 n floating point operations (arithmetic, comparison, and absolute value) and approximately 3 n bitwise operations. The final error bound with a 6-word reproducible accumulator and our default settings can be up to 2 29 times smaller than the error bound for conventional (recursive) summation on ill-conditioned double-precision inputs.

Computer Science↗

DOE OSTI2025

Optimal strategies for optical quantum memories using long-lived noble-gas spins

Nuclear spins of noble gases exhibit exceptionally long coherence times and can potentially serve as a long-lived storage medium for quantum information. We analyze and compare the performance of two mechanisms for mapping the quantum state of light onto the collective spin state of noble gases. The first mechanism utilizes collisional exchange with the electronic spin state of metastable noble-gas atoms, while the second relies on spin-exchange collisions with ground-state alkali-metal atoms. We describe the operation of an optical quantum memory relying on these two mechanisms using a compact model and study strategies that optimize the memory storage efficiency. Through numerical simulations, we identify optimal sequences for storing optical signals with different signal bandwidths and electronic spin relaxation rates. This work highlights the qualitative difference between the two approaches for using noble gases as long-lived quantum memories at noncryogenic conditions and outlines the regimes in which they are expected to be efficient.

atomic ensemble↗

DOE OSTI2023

Fast tree-based algorithms for DBSCAN for low-dimensional data on GPUs

DBSCAN is a well-known density-based clustering algorithm to discover arbitrary shape clusters. While conceptually simple in serial, the algorithm is challenging to efficiently parallelize on manycore GPU architectures. Common pitfalls, such as asynchronous range query calls, result in high thread execution divergence in many implementations. In this paper, we propose a new framework for GPU-accelerated DBSCAN, and describe two tree-based algorithms within that framework. Both algorithms fuse the search for neighbors with updating cluster information, but differ in their treatment of dense regions of the data. We show that the time taken to compute clusters is at most twice that of determination of the neighbors. We compare the proposed algorithms with existing CPU and GPU implementations, and demonstrate their competitiveness and performance using a fast traversal structure (bounding volume hierarchy) for low dimensional data. We also show that the memory usage can be reduced by processing object neighbors dynamically without storing them.

Prokopenko, Andrey↗

DOE OSTI2024

Modern deep neural networks for Direct Normal Irradiance forecasting: A classification approach

The escalating energy demand and the adverse environmental impacts of fossil-fuel use necessitate a shift towards cleaner and renewable alternatives. Concentrated Solar Power (CSP) technology emerges as a promising solution, offering a carbon-free alternative for power generation. The efficiency and profitability of CSP depend on the Direct Normal Irradiance (DNI) component of solar radiation; hence, accurate DNI forecasting can help optimize CSP plants’ operations and performance. The unpredictable nature of weather phenomena, particularly cloud cover, introduces uncertainty into DNI projections. Existing DNI forecasting models use meteorological factors, which are both challenging to estimate numerically over short prediction windows and expensive to model through data at a sufficiently high spatial and temporal resolution. This research addresses the challenge by presenting a novel approach that formulates DNI prediction as a multi-class classification problem, departing from conventional regression-based methods. The primary objective of this classification framework is to identify optimal periods aligning with specific operational thresholds for CSP plants, contributing to enhanced dispatch optimization strategies. We model the DNI classification problem using four advanced deep neural networks – rectified linear unit (ReLU) networks, 1D residual networks (ResNets), bidirectional long short-term memory (BiLSTM) networks, and transformers – achieving accuracies up to 93.5% without requiring meteorological parameters.

14 SOLAR ENERGY↗

DOE OSTI2022

Unified Language Frontend for Physic-Informed AI/ML

Artificial intelligence and machine learning (AI/ML) are becoming important tools for scientific modeling and simulation as in several other fields such as image analysis and natural language processing. ML techniques can leverage the computing power available in modern systems and reduce the human effort needed to configure experiments, interpret and visualize results, draw conclusions from huge quantities of raw data, and build surrogates for physics based models. Domain scientists in fields like fluid dynamics, microelectronics and chemistry can automate many of their most difficult and repetitive tasks or improve the design times by use of the faster ML-surrogates. However, modern ML and traditional scientific highperformance computing (HPC) tend to use completely different software ecosystems. While ML frameworks like PyTorch and TensorFlow provide Python APIs, most HPC applications and libraries are written in C++. Direct interoperability between the two languages is possible but is tedious and error-prone. In this work, we show that a compiler-based approach can bridge the gap between ML frameworks and scientific software with less developer effort and better efficiency. We use the MLIR (multi-level intermediate representation) ecosystem to compile a pre-trained convolutional neural network (CNN) in PyTorch to freestanding C++ source code in the Kokkos programming model. Kokkos is a programming model widely used in HPC to write portable, shared-memory parallel code that can natively target a variety of CPU and GPU architectures. Our compiler-generated source code can be directly integrated into any Kokkosbased application with no dependencies on Python or cross-language interfaces.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

DOE OSTI2024

Energy Exascale Computational Fluid Dynamics Simulations With the Spectral Element Method

Development and application of the open-source GPU-based fluid-thermal simulation code, NekRS, are described. Time advancement is based on an efficient kth-order accurate timesplit formulation coupled with scalable iterative solvers. Spatial discretization is based on the high-order spectral element method (SEM), which affords the use of fast, low-memory, matrix-free operator evaluation. Further, recent developments include support for nonconforming meshes using overset grids and for GPU-based Lagrangian particle tracking. Results of large-eddy simulations of atmospheric boundary layers for wind-energy applications as well as extensive nuclear energy applications are presented.

42 ENGINEERING↗

DOE OSTI2022

Position Papers for the ASCR Workshop on the Management and Storage of Scientific Data

The purpose of this workshop is to identify priority research directions in the area of data management for high-performance and scientific computing above and beyond HPC’s traditional "the parallel file system is the data-management system" model. Supporting the breadth of the DOE mission, including the explosion of AI uses and the growing needs of experimental and observational science, motivates revisiting our assumptions about data management. There are many facets of this topic to explore including: (1) Interfaces for accessing data that resides on traditional persistent storage as well as memory devices; (2) Storage-system architecture design that supports scientific workflows on varied hierarchical storage and networking devices; (3) Devising metadata management infrastructure to support FAIR principles (Findability, Accessibility, Interoperability, and Reusability); (4) Capturing provenance information about scientific data; (5) Utilizing AI to learn I/O patterns of emerging workloads for efficient data management; (6) Providing data management support for AI and complex workflows; and (7) Understanding the overlap between traditional storage systems and I/O (SSIO) efforts and data management. While the program committee has identified these topics as important areas for discussion, we welcome position papers from the community that propose additional topics of interest for discussion at the workshop. The workshop agenda will include breakout sessions for discussing these and selected topic areas to inform priority research directions for data management for high-performance and scientific computing.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2022

Report for the ASCR Workshop on the Management and Storage of Scientific Data

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

An immersed interface method for microstructure-scale electrochemical battery models: numerical formulation and performance portable implementation

We present the numerical formulation, verification, and performance portable implementation of an immersed interface method for microstructure scale electrochemical modeling of batteries. The innovation in this approach is the resolution of chemical species and electrostatic potential discontinuities at active interfaces without the use of interface conforming unstructured grids. A unified formulation on Cartesian grids for all domains (electrodes and electrolyte) is used with interfacial flux conditions applied using volume fraction or “color” function gradients. We have developed one dimensional and two dimensional test cases with analytic solutions for electrochemical modeling using which we verified the consistency and accuracy of our scheme. Our solver is also validated against solutions from a macroscale model and an unstructured multi-subdomain solver for a full lithium ion cell. We then demonstrated the utility of our solver on an image-based complex battery electrode microstructure at high charging rate. Our technique also exhibits good scalability on distributed memory architectures using central processing units (CPU), with problem sizes up to 1.8 billion degrees of freedom and with 5400 ranks. Initial performance studies of our open-source performance portable solver showed about 70 times speed up using a graphics processing unit (GPU) compared to single compute core for a problem with 4 million cells.

25 ENERGY STORAGE↗

DOE OSTI2025

Forecasting Multi-Step-Ahead Street-Scale Nuisance Flooding using a seq2seq LSTM Surrogate Model for Real-Time Application in a Coastal-Urban City

In coastal-urban cities facing an elevated risk of nuisance flooding (by rain and tide) due to increased heavy rainfall, sea level rise, urbanization, and aging drainage systems, real-time flood forecasting at the street-scale can provide useful information to transportation decision-makers. Physics-Based Models (PBMs) that offer high accuracy come with high computational runtimes and costs that limit their application for real-time flood forecasting. To address this challenge, Machine Learning (ML) surrogate models trained from PBMs have been proposed to provide street-scale flood forecasts. Previous related studies have focused on using Long Short-Term Memory (LSTM) architectures to model hourly flood depth on streets. While LSTM models can capture input sequences effectively, they fall short in accurately preserving output sequences, limiting their suitability for multi-step-ahead forecasts. The seq2seq LSTM architecture offers a key advantage here by capturing the full sequence of input–output, making it potentially more suitable for multi-step-ahead flood forecasts compared to traditional LSTM models. However, seq2seq LSTM has not been tested for street-scale flood forecasting, particularly for rapidly fluctuating nuisance flooding events which require special attention to its temporal sequences. Hence, in this study, we applied the seq2seq LSTM model to explore multi-step-ahead street-scale nuisance flooding and compared its results to the traditional LSTM model as a benchmark model. LSTM and seq2seq LSTM surrogate models were applied to 22 flood-prone streets in Norfolk, Virginia, as a case study with a 4-hr (short-term) and 8-hr (long-term) lead time. The models were trained with environmental (rainfall and tide) and topographic (elevation, Topographic Wetness Index, and Depth-To-Water) features along with PBM-derived water depths for different storm events. The results demonstrated satisfactory performance of both LSTM and seq2seq LSTM surrogate models throughout the forecast period compared to the PBM. However, the seq2seq LSTM showed lower Mean Absolute Error (MAE)/ Root Mean Square Error (RMSE) and higher Nash–Sutcliffe Efficiency (NSE)/ correlation than the LSTM across most lead times, particularly for long-term forecasting due to its supremacy in handling both input–output sequences together, which is missing in the traditional LSTM. For example, in the long-term, the average RMSE ranges were 0.0268–0.0373 m for LSTM and 0.0226–0.0319 m for seq2seq LSTM, while in the short-term, they were 0.0263–0.0293 m and 0.0261–0.0283 m, respectively. Additionally, while both models exhibited similar performance in distinguishing flooded and non-flooded streets for flood depth ≥ 0.1 m, the seq2seq LSTM model demonstrated superior performance for higher flood depths (such as ≥ 0.2 m and ≥ 0.3 m). Once trained, inference took only 0.09 to 0.11 s (short-term) and 0.30 to 0.35 s (long-term) per storm event for the 22 streets, making the application highly suitable for real-time decision-making during nuisance flood events.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2022

Spectral function for 4 He using the Chebyshev expansion in coupled-cluster theory

Here, we compute spectral function for 4 He by combining coupled-cluster theory with an expansion of integral transforms into Chebyshev polynomials. Our method allows us to estimate the uncertainty of spectral reconstruction. The properties of the Chebyshev polynomials make the procedure numerically stable and considerably lower in memory usage than the typically employed Lanczos algorithm. We benchmark our predictions with other calculations in the literature and with electron-scattering data in the quasi-elastic peak. The spectral function formalism allows one to extend ab initio lepton-nucleus cross sections into the relativistic regime. This makes it a promising tool for modeling this process at higher-energy transfers. The results we present open the door for studies of heavier nuclei, important for the neutrino oscillation programs.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2021

Fully Convolutional Spatio-Temporal Models for Representation Learning in Plasma Science

We have trained a fully convolutional spatio-temporal model for fast and accurate representation learning in the challenging exemplar application area of fusion energy plasma science. The onset of major disruptions is a critically important fusion energy science issue that must be resolved for advanced tokamak plasmas such as the $25B burning plasma international thermonuclear experimental reactor (ITER) experiment. While a variety of statistical methods have been used to address the problem of tokamak disruption prediction and control, recent approaches based on deep learning have proven particularly compelling. In the present paper, we introduce further improvements to the fusion recurrent neural network (FRNN) software suite, which delivered cross-machine disruption predictions with unprecedented accuracy using a large database of experimental signals from two major tokamaks. Up to now, FRNN was based on the long short-term memory (LSTM) variant of recurrent neural networks to leverage the temporal information in the data. Here, we implement and apply the "temporal convolutional neural network (TCN)" architecture to the time-dependent input signals. Furthermore, this allows highly optimized convolution operations to carry the majority of the computational load of training, thus enabling a reduction in training time, and the effective use of high-performance computing resources for hyperparameter tuning. At the same time, the TCN-based architecture achieves better predictive performance when compared with the LSTM architecture for various tasks for a representative fusion database.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DOE OSTI2022

Three-dimensional simulations of reshocked inclined Richtmyer-Meshkov instability: Effects of initial perturbations

The effect of initial perturbations on the evolution of the inclined Richtmyer-Meshkov turbulent mixing layer before and after reshock initiated by a shock wave with Mach number 1.55 is investigated through three-dimensional (3D) simulations using the flash code. The 3D simulations aim to reproduce both predominantly single-mode and multimode interfaces between light and heavy gases (N 2 -CO 2 , Atwood number, A≈0.22; amplitude to wavelength ratio of 0.088) which were created in an inclined shock tube facility to analyze the effects of initial conditions on mixing development in the entire flow field. The two-dimensional center slices of 3D simulations are compared with the experimental results to validate the computational code. Mixing width, mixed mass, mixed-mass thickness, and circulation in addition to concentration fields are shown to be in good agreement with the experimental data. The three-dimensional density and vorticity fields are first presented to qualitatively describe the flow behavior before and after reshock. Several measured density/velocity-related quantities indicate that the growth of the mixing material is strongly dependent on initial conditions. Before reshock and at early times after reshock, flow is clearly maintaining the memory of initial perturbations. However, at late time after reshock, although the large wavelength feature still dominates the flow motion, and the morphology of the two different interfaces indicates several differences, by breakdown of large-scale coherent structures to much finer scales, the memory of small scales of the multimode initial perturbation is not as clear as pre-reshock. Regarding three-dimensionality of the flow, before reshock in the multimode case, the baroclinic vorticity production, circulation, turbulent kinetic energy, and turbulent mass flux suggest that the small-scale roll-up features along the large inclined wavelength quickly evolves in all three dimensions. The coherent vortex tubes break down to smaller wormlike vortex structures, and turbulent fluctuations in the out-of-plane dimension are comparable to the spanwise direction. After reshock, this three-dimensionality of mixing growth was observed in the flow for both initial conditions. The results of this work represent a significant extension of previous computational studies performed on this specific topic. A different code with a different numerical method is validated through comparison with the experimental data. The initial perturbations are directly measured from the experimental results. Moreover, the entire three-dimensional experimental shock tube domain is simulated, and more quantities are investigated to understand the mixing mechanism and instability evolution in all three dimensions.

42 ENGINEERING↗

DOE OSTI2020

Time-series learning of latent-space dynamics for reduced-order model closure

In this work, we study the performance of long short-term memory networks (LSTMs) and neural ordinary differential equations (NODEs) in learning latent-space representations of dynamical equations for an advection-dominated problem given by the viscous Burgers equation. Our formulation is devised in a nonintrusive manner with an equation-free evolution of dynamics in a reduced space with the latter being obtained through a proper orthogonal decomposition. In addition, we leverage the sequential nature of learning for both LSTMs and NODEs to demonstrate their capability for closure in systems that are not completely resolved in the reduced space. We assess our hypothesis for two advection-dominated problems given by the viscous Burgers equation. We observe that both LSTMs and NODEs are able to reproduce the effects of the absent scales for our test cases more effectively than does intrusive dynamics evolution through a Galerkin projection. This result empirically suggests that time-series learning techniques implicitly leverage a memory kernel for coarse-grained system closure as is suggested through the Mori–Zwanzig formalism.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

LibraryX: A Framework for Cross-Library-Call Optimization

Scientific applications utilize performance libraries as a software engineering concept: these libraries encapsulate important and well-understood (mathematical) operations, allow for reuse, and are implemented and tuned by experts. Domain scientists then implement complex algorithms based on these domainspecific libraries. While individual library calls are optimized, larger performance gains across sequences of calls—sometimes spanning multiple libraries—are often unrealized, forcing a trade-off between performance and implementation complexity.To overcome this issue, we propose LibraryX, an approach and a system that allows for cross-library-call optimization even when library calls stem from multiple performance libraries. LibraryX annotates library calls with semantic information and optimizes entire directed acyclic graphs (DAGs) of calls dynamically using the SPIRAL code generation system. We demonstrate its effectiveness across a range of memory bound workloads, achieving significant speedups on Nvidia, AMD, and Intel accelerators compared to code using native libraries without cross-call optimization.

Rao, Sanil [Carnegie Mellon University,Department ↗

DOE OSTI2023

Newly Released Capabilities in the Distributed-Memory SuperLU Sparse Direct Solver

We present the new features available in the recent release of SuperLU_DIST, Version 8.1.1. SuperLU_DIST is a distributed-memory parallel sparse direct solver. The new features include (1) a 3D communication-avoiding algorithm framework that trades off inter-process communication for selective memory duplication, (2) multi-GPU support for both NVIDIA GPUs and AMD GPUs, and (3) mixed-precision routines that perform single-precision LU factorization and double-precision iterative refinement. Apart from the algorithm improvements, we also modernized the software build system to use CMake and Spack package installation tools to simplify the installation procedure. Throughout the article, we describe in detail the pertinent performance-sensitive parameters associated with each new algorithmic feature, show how they are exposed to the users, and give general guidance of how to set these parameters. We illustrate that the solver’s performance both in time and memory can be greatly improved after systematic tuning of the parameters, depending on the input sparse matrix and underlying hardware.

97 MATHEMATICS AND COMPUTING↗