Search NASA⌕ Search

SEARCH · Search NASA

Results for “near memory computing”

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 145 records · Page 8

Computational and Experimental Simulation of Static Memory Cells of Submicron Microcircuits under the Effect of Neutron Fluxes

The simulation of reversible single events in test samples of static memory microcircuits with design norms of 0.5, 0.35, 0.25, and 0.1 μm under the effect of neutron fluxes with various energies is performed. It is shown theoretically and experimentally that reversible single events can occur in modern microelectronics and nanoelectronics products under the effect of a fission-spectrum neutron flux caused by the passage of primary recoil atoms and nuclear reaction products along the microcircuit surface perpendicularly to the electric current lines in the near-drain transistor area. A series of irradiation experiments of static memory circuits with design norms of 0.35 μm is interpreted based on the proposed model.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Probabilistic Nanomagnetic Memories for Uncertain and Robust Machine Learning

This project evaluated the use of emerging spintronic memory devices for robust and efficient variational inference schemes. Variational inference (VI) schemes, which constrain the distribution for each weight to be a Gaussian distribution with a mean and standard deviation, are a tractable method for calculating posterior distributions of weights in a Bayesian neural network such that this neural network can also be trained using the powerful backpropagation algorithm. Our project focuses on domain-wall magnetic tunnel junctions (DW-MTJs), a powerful multi-functional spintronic synapse design that can achieve low power switching while also opening the pathway towards repeatable, analog operation using fabricated notches. Our initial efforts to employ DW-MTJs as an all-in-one stochastic synapse with both a mean and standard deviation didn’t end up meeting the quality metrics for hardware-friendly VI. In the future, new device stacks and methods for expressive anisotropy modification may make this idea still possible. However, as a fall back that immediately satisfies our requirements, we invented and detailed how the combination of a DW-MTJ synapse encoding the mean and a probabilistic Bayes-MTJ device, programmed via a ferroelectric or ionically modifiable layer, can robustly and expressively implement VI. This design includes a physics-informed small circuit model, that was scaled up to perform and demonstrate rigorous uncertainty quantification applications, up to and including small convolutional networks on a grayscale image classification task, and larger (Residual) networks implementing multi-channel image classification. Lastly, as these results and ideas all depend upon the idea of an inference application where weights (spintronic memory states) remain non-volatile, the retention of these synapses for the notched case was further interrogated. These investigations revealed and emphasized the importance of both notch geometry and anisotropy modification in order to further enhance the endurance of written spintronic states. In the near future, these results will be mapped to effective predictions for room temperature and elevated operation DW-MTJ memory retention, and experimentally verified when devices become available.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Computational Modeling and Experimental Characterization of Martensitic Transformations in Nicoal for Self-Sensing Materials

Fundamental changes to aero-vehicle management require the utilization of automated health monitoring of vehicle structural components. A novel method is the use of self-sensing materials, which contain embedded sensory particles (SP). SPs are micron-sized pieces of shape-memory alloy that undergo transformation when the local strain reaches a prescribed threshold. The transformation is a result of a spontaneous rearrangement of the atoms in the crystal lattice under intensified stress near damaged locations, generating acoustic waves of a specific spectrum that can be detected by a suitably placed sensor. The sensitivity of the method depends on the strength of the emitted signal and its propagation through the material. To study the transition behavior of the sensory particle inside a metal matrix under load, a simulation approach based on a coupled atomistic-continuum model is used. The simulation results indicate a strong dependence of the particle's pseudoelastic response on its crystallographic orientation with respect to the loading direction and suggest possible ways of optimizing particle sensitivity. The technology of embedded sensory particles will serve as the key element in an autonomous structural health monitoring system that will constantly monitor for damage initiation in service, which will enable quick detection of unforeseen damage initiation in real-time and during onground inspections.

Wallace, T. A.↗

Coupled full core neutron transport/CFD simulations of pressurized water reactors

Recently as part of the CASL project, a capability to perform 3D whole-core coupled neutron transport and computational fluid dynamics (CFD) calculations was demonstrated. This work uses the 2D/1D transport code DeCART and the commercial CFD code STAR-CCM+. It builds on previous CASL work demonstrating coupling for smaller spatial domains. The coupling methodology is described along with the problem simulated and results are presented for fresh hot full power conditions. An additional comparison is made to an equivalent model that uses lower order T/H feedback to assess the importance and cost of high fidelity feedback to the neutronics problem. A simulation of a quarter core Combustion Engineering (CE) PWR core was performed with the coupled codes using a Fixed Point Gauss-Seidel iteration technique. The total approximate calculation requirements are nearly 10,000 CPU hours and 1 TB of memory. The problem took 6 coupled iterations to converge. The CFD coupled model and low order T/H feedback model compared well for global solution parameters, with a difference in the critical boron concentration and average outlet temperature of 14 ppm B and 0.94 deg. C, respectively. Differences in the power distribution were more significant with maximum relative differences in the core-wide pin peaking factor (Fq) of 5.37% and average relative differences in flat flux region power of 11.54%. Future work will focus on analyzing problems more relevant to CASL using models with less approximations. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Electrochemical Random-Access Memory: Progress, Perspectives, and Opportunities

Non-von Neumann computing using neuromorphic systems based on analogue synaptic and neuronal elements has emerged as a potential solution to tackle the growing need for more efficient data processing, but progress toward practical systems has been stymied due to a lack of materials and devices with the appropriate attributes. Recently, solid state electrochemical ion-insertion, also known as electrochemical random access memory (ECRAM) has emerged as a promising approach to realize the needed device characteristics. ECRAM is a three terminal device that operates by tuning electronic conductance in functional materials through solid-state electrochemical redox reactions. This mechanism can be considered as a gate-controlled bulk modulation of dopants and/or phases in the channel. Early work demonstrating that ECRAM can achieve nearly ideal analogue synaptic characteristics has sparked tremendous interest in this approach. More recently, the realization that electrochemical ion insertion can be used to tune the electronic properties of many types of materials including transition metal oxides, layered two-dimensional materials, organic and coordination polymers, and that the changes in conductance can span orders of magnitude has further attracted interest in ECRAM as the basis for analogue synaptic elements for inference accelerators as well as for dynamical devices that can emulate a wide range of neuronal characteristics for implementation in analogue spiking neural networks. At its core, ECRAM shares many fundamental aspects with rechargeable batteries, where ion insertion materials are used extensively for their ability to reversibly store charge and energy. Computing applications, however, present drastically different requirements: systems will require many millions of devices, scaled down to tens of nanometers, all while achieving reliable electronic-state tuning at scaled-up rates and endurances, and with minimal energy dissipation and noise. Further, in this review, we discuss the history, basic concepts, recent progress, as well as the challenges and opportunities for different types of ECRAM, broadly grouped by their primary mobile ionic charge carrier, including Li, protons, and oxygen vacancies.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Spacecraft computer technology at Southwest Research Institute

Southwest Research Institute (SwRI) has developed and delivered spacecraft computers for a number of different near-Earth-orbit spacecraft including shuttle experiments and SDIO free-flyer experiments. We describe the evolution of the basic SwRI spacecraft computer design from those weighing in at 20 to 25 lb and using 20 to 30 W to newer models weighing less than 5 lb and using only about 5 W, yet delivering twice the processing throughput. Because of their reduced size, weight, and power, these newer designs are especially applicable to planetary instrument requirements. The basis of our design evolution has been the availability of more powerful processor chip sets and the development of higher density packaging technology, coupled with more aggressive design strategies in incorporating high-density FPGA technology and use of high-density memory chips. In addition to reductions in size, weight, and power, the newer designs also address the necessity of survival in the harsh radiation environment of space. Spurred by participation in such programs as MSTI, LACE, RME, Delta 181, Delta Star, and RADARSAT, our designs have evolved in response to program demands to be small, low-powered units, radiation tolerant enough to be suitable for both Earth-orbit microsats and for planetary instruments. Present designs already include MIL-STD-1750 and Multi-Chip Module (MCM) technology with near-term plans to include RISC processors and higher-density MCM's. Long term plans include development of whole-core processors on one or two MCM's.

Shirley, D. J.↗

Towards large-scale calculations with State-Specific Multireference Coupled Cluster methods: Studies on dodecane, naphthynes, and polycarbenes

Multireference Coupled Cluster (MRCC) methods are essential for accurate descriptions of molecular systems that have nearly degenerate states. A number of these methods have been developed over the last three decades but their use has been hampered by their high computational cost. The development of highly scalableMRCC codes can significantly alleviate this problem. Our recent work on the parallelization of the Brillouin-Wigner MRCC implementation has clearly demonstrated that taking advantage of large computational resources can enable the MRCC calculations on chemically relevant molecules within practical time-to-solution and memory per core constraints. In this letter we report on a parallel implementation of the Mukherjee’s state-specific MRCC model with single and double excitations (MR MkCCSD). In particular the abstraction of a methyl-group from dodecane, the singlet-triplet splitting in naphthyne isomers, and the singlet-high-spin splitting in polycarbenes are considered as demonstrations of systems with significant multireference character and of challenging size that currently can be handled with the MR MkCCSD formalism.

74 ATOMIC AND MOLECULAR PHYSICS↗

Julia as a unifying end-to-end workflow language on the Frontier exascale system

We evaluate Julia as a single language and ecosystem paradigm powered by LLVM to develop workflow components for high-performance computing. We run a Gray-Scott, 2-variable diffusion-reaction application using a memory-bound, 7-point stencil kernel on Frontier, the US Department of Energy’s first exascale supercomputer. We evaluate the performance, scaling, and trade-offs of (i) the computational kernel on AMD’s MI250x GPUs, (ii) weak scaling up to 4,096 MPI processes/GPUs or 512 nodes, (iii) parallel I/O writes using the ADIOS2 library bindings, and (iv) Jupyter Notebooks for interactive analysis. Results suggest that although Julia generates a reasonable LLVM-IR, a nearly 50% performance difference exists vs. native AMD HIP stencil codes when running on the GPUs. As expected, we observed near-zero overhead when using MPI and parallel I/O bindings for system-wide installed implementations. Consequently, Julia emerges as a compelling high-performance and high-productivity workflow composition language, as measured on the fastest supercomputer in the world.

Godoy, William↗

Achieving High Efficiency in Reduced Order Modeling for Large Scale Polycrystal Plasticity Simulations

Reduced order models for the nonlinear response of heterogeneous microstructures typically require a construction (or training) stage to build the reduced order basis. In this manuscript, an efficient model construction strategy for the eigenstrain homogenization method (EHM) is presented. The proposed strategy relies on a parallel, element-by-element, conjugate gradient solver. Near linear scaling has been achieved with respect to the number of degrees of freedom used to resolve the microstructure. Linear scaling with respect to the number of pre-analyses required to construct the reduced order model (ROM) follows from the EHM formulation. Furthermore, a parallel implementation for fast evaluation of the constructed ROM has been developed using shared memory parallelization. It has been shown that for large microstructures with ≈ 10,000 grains, the total computational cost of evaluating the nonlinear response of a polycrystal could be reduced by approximately an order of magnitude using 32 cores with respect to serial ROM simulation. The present methodology has been verified using an additively manufactured polycrystalline microstructure of a nickel-based superalloy, Inconel 625. The capability of the developed framework to construct a ROM for such large microstructures, as well as the ability of the ROM to predict average and local quantities of interest has been demonstrated.

microscale↗

Computing material volume fractions on a superimposed mesh as applied to Monte Carlo particle transport simulations

Here, we present a newly implemented ray tracing algorithm in OpenMC for efficiently computing material volume fractions on superimposed meshes in complex geometries. By firing rays along each coordinate direction through the geometry, the approach accumulates track-length data in each mesh element, thereby determining the fractional composition of each material. Scaling studies on three different models—a random tetrahedra configuration, the Frascati Neutron Generator ITER dose rate benchmark, and a stellarator design—show excellent parallel performance, with nearly linear speedup on modern multi-threaded and distributed-memory systems. An analysis of the residual error relative to high-resolution reference solutions demonstrated that under optimal conditions it decreases as 1/R, where R is the number of rays fired, making it straightforward to achieve user-prescribed accuracy. This new functionality enables practical, mesh-based approaches for detailed nuclear analyses in production Monte Carlo workflows without resorting to expensive, fully conformal or unstructured meshing.

Monte Carlo↗

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↗

Sage: parallel semi-asymmtric graph algorithms for NVRAMs

Non-volatile main memory (NVRAM) technologies provide an attractive set of features for large-scale graph analytics, including byte-addressability, low idle power, and improved memory-density. NVRAM systems today have an order of magnitude more NVRAM than traditional memory (DRAM). NVRAM systems could therefore potentially allow very large graph problems to be solved on a single machine, at a modest cost. However, a significant challenge in achieving high performance is in accounting for the fact that NVRAM writes can be much more expensive than NVRAM reads. In this paper, we propose an approach to parallel graph analytics using the Parallel Semi-Asymmetric Model (PSAM), in which the graph is stored as a read-only data structure (in NVRAM), and the amount of mutable memory is kept proportional to the number of vertices. Similar to the popular semi-external and semi-streaming models for graph analytics, the PSAM approach assumes that the vertices of the graph fit in a fast read-write memory (DRAM), but the edges do not. In NVRAM systems, our approach eliminates writes to the NVRAM, among other benefits. To experimentally study this new setting, we develop Sage, a parallel semi-asymmetric graph engine with which we implement provably-efficient (and often work-optimal) PSAM algorithms for over a dozen fundamental graph problems. We experimentally study Sage using a 48--core machine on the largest publicly-available real-world graph (the Hyperlink Web graph with over 3.5 billion vertices and 128 billion edges) equipped with Optane DC Persistent Memory, and show that Sage outperforms the fastest prior systems designed for NVRAM. Importantly, we also show that Sage nearly matches the fastest prior systems running solely in DRAM, by effectively hiding the costs of repeatedly accessing NVRAM versus DRAM.

97 MATHEMATICS AND COMPUTING↗

Running SW4 On New Commodity Technology Systems (CTS-1) Platform

We have recently been running earthquake ground motion simulations with SW4 on the new capacity computing systems, called the Commodity Technology Systems - 1 (CTS-1) at Lawrence Livermore National Laboratory (LLNL). SW4 is a fourth order time domain finite difference code developed by LLNL and distributed by the Computational Infrastructure for Geodynamics (CIG). SW4 simulates seismic wave propagation in complex three-dimensional Earth models including anelasticity and surface topography. We are modeling near-fault earthquake strong ground motions for the purposes of evaluating the response of engineered structures, such as nuclear power plants and other critical infrastructure. Engineering analysis of structures requires the inclusion of high frequencies which can cause damage, but are often difficult to include in simulations because of the need for large memory to model fine grid spacing on large domains.

58 GEOSCIENCES↗

Developments and Validations of Fully Coupled CFD and Practical Vortex Transport Method for High-Fidelity Wake Modeling in Fixed and Rotary Wing Applications

A novel Computational Fluid Dynamics (CFD) coupling framework using a conventional Reynolds-Averaged Navier-Stokes (BANS) solver to resolve the near-body flow field and a Particle-based Vorticity Transport Method (PVTM) to predict the evolution of the far field wake is developed, refined, and evaluated for fixed and rotary wing cases. For the rotary wing case, the RANS/PVTM modules are loosely coupled to a Computational Structural Dynamics (CSD) module that provides blade motion and vehicle trim information. The PVTM module is refined by the addition of vortex diffusion, stretching, and reorientation models as well as an efficient memory model. Results from the coupled framework are compared with several experimental data sets (a fixed-wing wind tunnel test and a rotary-wing hover test).

Anusonti-Inthra, Phuriwat↗

Accelerating String Set Matching in FPGA Hardware for Bioinformatics Research

Background: This paper describes techniques for accelerating the performance of the string set matching problem with particular emphasis on applications in computational proteomics. The process of matching peptide sequences against a genome translated in six reading frames is part of a proteogenomic mapping pipeline that is used as a case-study. The Aho-Corasick algorithm is adapted for execution in field programmable gate array (FPGA) devices in a manner that optimizes space and performance. In this approach, the traditional Aho-Corasick finite state machine (FSM) is split into smaller FSMs, operating in parallel, each of which matches up to 20 peptides in the input translated genome. Each of the smaller FSMs is further divided into five simpler FSMs such that each simple FSM operates on a single bit position in the input (five bits are sufficient for representing all amino acids and special symbols in protein sequences). Results: This bit-split organization of the Aho-Corasick implementation enables efficient utilization of the limited random access memory (RAM) resources available in typical FPGAs. The use of onchip RAM as opposed to FPGA logic resources for FSM implementation also enables rapid reconfiguration of the FPGA without the place and routing delays associated with complex digital designs. Conclusion: Experimental results show storage efficiencies of over 80% for several data sets. Furthermore, the FPGA implementation executing at 100 MHz is nearly 20 times faster than an implementation of the traditional Aho-Corasick algorithm executing on a 2.67 GHz workstation.

59 BASIC BIOLOGICAL SCIENCES↗

Randomized Projection for Rank-Revealing Matrix Factorizations and Low-Rank Approximations

Rank-revealing matrix decompositions provide an essential tool in spectral analysis of matrices, including the Singular Value Decomposition (SVD) and related low-rank approximation techniques. QR with Column Pivoting (QRCP) is usually suitable for these purposes, but it can be much slower than the unpivoted QR algorithm. For large matrices, the difference in performance is due to increased communication between the processor and slow memory, which QRCP needs in order to choose pivots during decomposition. Our main algorithm, Randomized QR with Column Pivoting (RQRCP), uses randomized projection to make pivot decisions from a much smaller sample matrix, which we can construct to reside in a faster level of memory than the original matrix. This technique may be understood as trading vastly reduced communication for a controlled increase in uncertainty during the decision process. Furthermore, for rank-revealing purposes, the selection mechanism in RQRCP produces results that are the same quality as the standard algorithm, but with performance near that of unpivoted QR (often an order of magnitude faster for large matrices). Additionally, we also propose two formulas that facilitate further performance improvements. The first efficiently updates sample matrices to avoid computing new randomized projections. The second avoids large trailing updates during the decomposition in truncated low-rank approximations. Our truncated version of RQRCP also provides a key initial step in our truncated SVD approximation, TUXV. These advances open up a new performance domain for large matrix factorizations that will support efficient problem-solving techniques for challenging applications in science, engineering, and data analysis.

97 MATHEMATICS AND COMPUTING↗

Simulating many-engine spacecraft: Exceeding 1 quadrillion degrees of freedom via information geometric regularization

We present an optimized implementation of the recently proposed information geometric regularization (IGR) for unprecedented scale simulation of compressible fluid flows applied to multi-engine spacecraft boosters. We improve upon state-of-the-art computational fluid dynamics (CFD) techniques in terms of computational cost, memory footprint, and energy-to-solution metrics. Unified memory on coupled CPU–GPU or APU platforms increases problem size with negligible overhead. Mixed half/single-precision storage and computation are used on well-conditioned numerics. We simulate flow at 200 trillion grid points and 1 quadrillion degrees of freedom, exceeding the current record by a factor of 20. A factor of 4 wall-time speedup is achieved over optimized baselines. Ideal weak scaling is observed on OLCF Frontier, LLNL El Capitan, and CSCS Alps using the full systems. Strong scaling is near ideal at extreme conditions, including 80% efficiency on CSCS Alps with an 8 node baseline and stretching to the full system.

Wilfong, Benjamin [Georgia Institute of Technology↗

Fast Machine Learning Lidar Surrogate Simulator: Pristine Clear Sky

The simulations of lidar signals and retrievals rely on a range of optic-physical models, such as radiative transfer models, particle scattering and absorption models, along with the output data from atmospheric physical models. Integrating these different models to represent signals of a lidar system is computationally expensive, and performing backward retrievals can be complex and ambiguous. However, with the advantages of Machine Learning, there is a new potential for building effective lidar signal database linked to corresponding atmospheric profiles. For this project, we are developing a fast pre-trained neural network as the lidar surrogate simulator using simulated data for a CALIPSO-like lidar (355 nm, 532nm, and 1064nm), and a CO2 differential absorption lidar (DIAL) near 1571nm. Specifically, we utilize a long short-term memory (LSTM) model to map the relationships between atmospheric profiles (pressure, temperature, air density and CO2 mixing ratio) and lidar signals. This approach allows us to build machine learning based simulators that can reconstruct lidar signals at specific bands from MERRA reanalysis data, and perform retrievals of atmospheric profiles using lidar signals at various wavelengths. As a first step, the results show the potential of this method to establish a foundational model for sensor signals. This model offers the promise of enabling both accurate predictions and rapid retrievals, providing a more efficient approach to signal processing and analysis.

Shan Zeng↗