Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “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 433 records · Page 24
All sourcesNASA NTRSDOE OSTI
NASA NTRSDate not supplied

Performance Optimization Methods for a Memory-Bound, Unstructured-Grid CFD Application on Massively Parallel GPU Platforms

Computational performance of the FUN3D unstructured-grid computational fluid dynamics (CFD) application on massively parallel GPU environments is memory-bound and highly dependent upon efficient reads from and atomic updates to the irregular cell-, edge-, and node-based data structures. In this talk, we present recent efforts into optimizing select performance-critical kernels on NVIDIA Tesla V100 and A100 GPUs and AMD CDNA MI100 GPUs. A novel use of L2 cache residency controls and asynchronous loads into on-chip shared memory are explored on the A100 GPU for the sparse iterative solver, which is dominated by mixed-precision, sparse matrix vector multiplication. Demonstrations show that these methods improve global memory bandwidth utilization by 13.5% on the A100 GPU. Several techniques are also presented that use registers and/or shared memory to facilitate array transposition and aggregation which combine to reduce the frequency and increase the cache efficiency of floating-point atomic updates to the irregular data structures. These methods are demonstrated to improve the kernel throughput by nearly 500% on select kernels on the AMD MI100 over atomic updates directly to global memory. Overall, both V100 and A100 GPUs outperformed the MI100 GPU on kernels dominated by double-precision atomic updates; however, the techniques demonstrated here reduced the performance gap and improved the MI100 performance.

GPU CPU unstructured CFD memory↗
NASA NTRSDate not supplied

Design and Analysis of Shape Memory Spring Tires for Martian and Lunar Rover Vehicles

Shape memory alloys (SMAs) have played an important role in various innovative engineering and medical applications, such as aerospace actuators, vibration damping devices, and coronary stents. In applications, shape memory alloys are commonly utilized in two fundamentally different ways: (i) making use of the superelasticity/ pseudoelasticity (SE/PE) phenomena, as in applications in biomedical engineering, and (ii) taking advantages of the shape memory effect (SME), as is used for actuators. Their ability to act in such vastly different capacities is mainly due to their unique capability to recover large amounts of deformation produced by either applied stresses or temperature changes. One recent emerging application in use of SMAs has been in the area of non-pneumatic tire designs for Martian or Lunar roving vehicles. These vehicles require tires that are capable of traversing rugged terrain while withstanding extreme temperatures and atmospheric conditions. Inspired by the flexible wire mesh tires used on three Lunar Roving Vehicle (LRV) missions to the Moon on Apollo 15, 16, and 17, a new compliant tire technology was developed by the NASA Glenn Research Center (GRC) and Goodyear Tire & Rubber, known as the Spring Tire. The Spring Tire consists of several hundred coiled springs woven into a flexible mesh and formed into the shape of a tire. Like the LRV wire mesh tires, the original Spring Tires were made from spring steel and were prone to permanent deformation when undergoing high localized loads. Later, a new iteration of this technology was invented, known as the ‘Superelastic Tire’. This new technology incorporated the use of superelastic SMA springs, which could effectively undergo approximately 30 times more reversible deformation than the steel spring. It also provided even greater durability and allowed for more flexibility in design, such as the use of other structural forms to reduce mass or increase load carrying capacity. Because of the unique nature of both the SMA material and the complex interactions between the springs, designing Spring Tires for a specific application requires extensive effort. Historically, design decisions have relied on full-scale empirical testing; however, this is very expensive and time consuming, especially when multiple iterations are needed. Therefore, developing a large-scale, robust, and predictive numerical model entailing complex spring interactions and the shape memory material behavior within a tire construct is the first essential step towards a successful design program. The current work focuses on implementation of the user-defined Shape Memory Alloy (SMA) model, otherwise known as SMA-GVIPs, in the Finite Element analysis (FEA) program ABAQUS for large-scale simulations of the GRC-developed Spring Tires made of SMA. The novelty of this work lies in the thorough, detail-oriented, and computationally efficient finite element analysis of full-scale SMA tires. A well-thought material characterization plan followed by model validation and a systemic sensitivity study on SMA tires has never been reported in the previous literature. The main objective of this study is to help the team improve and optimize the structural design of the SMA tires through in-depth numerical analysis and sensitivity studies. Various design variables (wire diameter, coil diameter, pitch, bead angle, and number of springs) were varied to study their influence on the global load-displacement response of the tire construct. A detailed investigation of the three-dimensional stress states was also carried out to enhance our understanding of the local changes as the tire goes through global deformation. It was concluded that a robust numerical model with a good predictive capability, together with a thoughtfully crafted sensitivity study can result in improved design iterations required to reach a desired tire performance while, significantly reducing manufacturing, labor and testing expenses. A summary of the Finite Element (FE) model construction will be presented together with a description of the user-defined SMA model, characterization process, experimental results, model validation, and numerical sensitivity study results.

shape memory alloys↗
DOE OSTI2016

Havens: Explicit Reliable Memory Regions for HPC Applications

Supporting error resilience in future exascale-class supercomputing systems is a critical challenge. Due to transistor scaling trends and increasing memory density, scientific simulations are expected to experience more interruptions caused by transient errors in the system memory. Existing hardware-based detection and recovery techniques will be inadequate to manage the presence of high memory fault rates. In this paper we propose a partial memory protection scheme based on region-based memory management. We define the concept of regions called havens that provide fault protection for program objects. We provide reliability for the regions through a software-based parity protection mechanism. Our approach enables critical program objects to be placed in these havens. The fault coverage provided by our approach is application agnostic, unlike algorithm-based fault tolerance techniques.

exascale computing↗
DOE OSTI2024

Nonvolatile electrochemical memory at 600°C enabled by composition phase separation

Silicon-based microelectronics are limited to ~150°C and therefore not suitable for the extremely high temperatures in aerospace, energy, and space applications. While wide-band-gap semiconductors can provide high-temperature logic, nonvolatile memory devices at high temperatures have been challenging. In this work, we develop a nonvolatile electrochemical memory cell that stores and retains analog and digital information at temperatures as high as 600°C. Through correlative scanning transmission electron microscopy, we show that this high-temperature information retention is a result of composition phase separation between the oxidized and reduced forms of amorphous tantalum oxide. This result demonstrates a memory concept that is resilient at extreme temperatures and reveals phase separation as the principal mechanism that enables nonvolatile information storage in these electrochemical memory cells.

42 ENGINEERING↗
DOE OSTI2022

A survey on processing-in-memory techniques: Advances and challenges

Processing-in-memory (PIM) techniques have gained much attention from computer architecture researchers, and significant research effort has been invested in exploring and developing such techniques. Increasing the research activity dedicated to improving PIM techniques will hopefully help deliver PIM’s promise to solve or significantly reduce memory access bottleneck problems for memory-intensive applications. We also believe it is imperative to track the advances made in PIM research to identify open challenges and enable the research community to make informed decisions and adjust future research directions. In this survey, we analyze recent studies that explored PIM techniques, summarize the advances made, compare recent PIM architectures, and identify target application domains and suitable memory technologies. We also discuss proposals that address unresolved issues of PIM designs (e.g., address translation/mapping of operands, workload analysis to identify application segments that can be accelerated with PIM, OS/runtime support, and coherency issues that must be resolved to incorporate PIM). We believe this work can serve as a useful reference for researchers exploring PIM techniques.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

New Total-Ionizing-Dose Resistant Data Storing Technique for NAND Flash Memory

This paper describes a new non-charge-based data storing technique in NAND flash memory called watermark that encodes read-only data in the form of physical properties of flash memory cells. Unlike traditional charge-based data storing method in flash memory, the proposed technique is resistant to total ionizing dose (TID) effects. To evaluate its resistance to irradiation effects, we analyze data stored in several commercial single-level-cell (SLC) flash memory chips from different vendors and technology nodes. These chips are irradiated using a Co-60 gamma-ray source array for up to 100 krad(Si) at Sandia National Laboratories. Experimental evaluation performed on a flash chip from Samsung shows that the intrinsic bit error rate (BER) of watermark increases from 0.8% for TID = 0 krad(Si) to 1% for TID = 100 krad(Si). Conversely, the BER of charge-based data stored on the same chip increases from 0% at TID = 0 krad(Si) to 1.5% at TID = 100 krad(Si). Overall, the results imply that the proposed technique may potentially offer significant improvements in data integrity relative to traditional charge-based data storage for very high radiation (TID > 100 krad(Si)) environments. These gains in data integrity relative to the charge-based data storage are useful in radiation-prone environments, but they come at the cost of increased write times and higher BERs before irradiation.

36 MATERIALS SCIENCE↗
DOE OSTI2017

Development of Conjugated Polymers for Memory Device Applications

This review summarizes the most widely used mechanisms in memory devices based on conjugated polymers, such as charge transfer, space charge traps, and filament conduction. In addition, recent studies of conjugated polymers for memory device applications are also reviewed, discussed, and differentiated based on the mechanisms and structural design. Moreover, the electrical conditions of conjugated polymers can be further fine-tuned by careful design and synthesis based on the switching mechanisms. The review also emphasizes and demonstrates the structure-memory properties relationship of donor-acceptor conjugated polymers for advanced memory device applications.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Role of α-Fetoprotein in Regulation of Proliferation and Functional Activity of Naïve T Cells and Immune Memory T Cells

We studied the role of native α-fetoprotein preparation in the regulation of proliferation and functional activity of naïve T cells and immune memory T cells in vitro. The study was carried out on separated fractions of naïve T cells (CD45RA{sup +}) and immune memory T cells (CD45R0{sup +}) incubated with α-fetoprotein under conditions of TCR activation. At the level of naïve T cells, α-fetoprotein in a concentration of 100 U/ml reduced the expression of CD28, but increased the expression of CD25, while at the level of immune memory T cells α-fetoprotein (50 and 100 U/ml) only suppressed the expression of CD25. No effects of α-fetoprotein on the proliferative status of the studied lymphocyte subpopulations and on the expression of CD71 (proliferation marker) by these cells were detected. Addition of α-fetoprotein in a concentration of 100 U/ml increased the level of IL-2 in naïve T cell culture supernatants, while production of IL-2 by memory T cells remained unchanged. These data demonstrated the priority aspects of regulation of the functional activities of naïve T cells and immune memory T cells.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2021

JTAG-based PLC memory acquisition framework for industrial control systems

In industrial control systems (ICS), programmable logic controllers (PLC) are the embedded devices that directly control and monitor critical industrial infrastructure processes such as nuclear plants and power grid stations. Cyberattacks often target PLCs to sabotage a physical process. A memory forensic analysis of a suspect PLC can answer questions about an attack, including compromised firmware and manipulation of PLC control logic code and I/O devices. Given physical access to a PLC, collecting forensic information from the PLC memory at the hardware-level is risky and challenging. It may cause the PLC to crash or hang since PLCs have proprietary, legacy hardware with heterogeneous architecture. This paper addresses this research problem and proposes a novel JTAG (Joint Test Action Group)-based framework, Kyros, for reliable PLC memory acquisition. Kyros systematically creates a JTAG profile of a PLC through hardware assessment, JTAG pins identification, memory map creation, and optimizing acquisition parameters. It also facilitates the community of interest (such as ICS owners, operators, and vendors) to develop the JTAG profiles of PLCs. Further, we present a case study of Kyros implementation over Allen-Bradley 1756-A10/B to help understand the framework's application on a real-world PLC used in industry settings. The sample PLC memory dumps are shared with the research community to facilitate further research.

Rais, Muhammad Haris↗
DOE OSTI2022

A second-order distributed memory parallel fast sweeping method for the Eikonal equation

The Eikonal equation is used to calculate wave propagation and distance fields, and due to its complexity requires numerical treatment for its solution. In this work, we present a second-order distributed memory parallel fast sweeping method. The second-order solution switches on a two-point stencil when two upwind points are available, and reverts to first-order otherwise. In all examples, the second-order method improves the solution over the first-order, allowing for significant savings in memory while achieving the same accuracy. Parallelization over distributed memory saw good weak scaling with optimal convergence. The computational time for second-order was approximately 2.5 times slower than first-order, where the largest amount of mesh points ran on 144 cores (512 GB) was ≈20 billion. The savings in memory from the second-order method combined with the distributed memory algorithm result in the ability to solve problems much larger than are possible with the serial first-order method.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Hong-Ou-Mandel interference of single-photon-level pulses stored in independent room-temperature quantum memories

Abstract Quantum repeater networks require independent absorptive quantum memories capable of storing and retrieving indistinguishable photons to perform high-repetition entanglement swapping operations. The ability to perform these coherent operations at room temperature is of prime importance for the realization of scalable quantum networks. We perform Hong-Ou-Mandel (HOM) interference between photonic polarization states and single-photon-level pulses stored and retrieved from two sets of independent room-temperature quantum memories. We show that the storage and retrieval of polarization states from quantum memories does not degrade the HOM visibility for few-photon-level polarization states in a dual-rail configuration. For single-photon-level pulses, we measure the HOM visibility with various levels of background in a single polarization, single-rail QM, and investigate its dependence on the signal-to-background ratio. We obtain an HOM visibility of 43%, compared to the 48% no-memory limit of our set-up. These results allow us to estimate a 33% visibility for polarization qubits under the same conditions. These demonstrations lay the groundwork for future applications using large-scale memory-assisted quantum networks.

Physics↗
DOE OSTI2014

Enhanced memory effect via quantum confinement in 16 nm InN nanoparticles embedded in ZnO charge trapping layer

In this work, the fabrication of charge trapping memory cells with laser-synthesized indium-nitride nanoparticles (InN-NPs) embedded in ZnO charge trapping layer is demonstrated. Atomic layer deposited Al{sub 2}O{sub 3} layers are used as tunnel and blocking oxides. The gate contacts are sputtered using a shadow mask which eliminates the need for any lithography steps. High frequency C-V{sub gate} measurements show that a memory effect is observed, due to the charging of the InN-NPs. With a low operating voltage of 4 V, the memory shows a noticeable threshold voltage (V{sub t}) shift of 2 V, which indicates that InN-NPs act as charge trapping centers. Without InN-NPs, the observed memory hysteresis is negligible. At higher programming voltages of 10 V, a memory window of 5 V is achieved and the V{sub t} shift direction indicates that electrons tunnel from channel to charge storage layer.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2014

Resummed memory kernels in generalized system-bath master equations

Generalized master equations provide a concise formalism for studying reduced population dynamics. Usually, these master equations require a perturbative expansion of the memory kernels governing the dynamics; in order to prevent divergences, these expansions must be resummed. Resummation techniques of perturbation series are ubiquitous in physics, but they have not been readily studied for the time-dependent memory kernels used in generalized master equations. In this paper, we present a comparison of different resummation techniques for such memory kernels up to fourth order. We study specifically the spin-boson Hamiltonian as a model system bath Hamiltonian, treating the diabatic coupling between the two states as a perturbation. A novel derivation of the fourth-order memory kernel for the spin-boson problem is presented; then, the second- and fourth-order kernels are evaluated numerically for a variety of spin-boson parameter regimes. We find that resumming the kernels through fourth order using a Padé approximant results in divergent populations in the strong electronic coupling regime due to a singularity introduced by the nature of the resummation, and thus recommend a non-divergent exponential resummation (the “Landau-Zener resummation” of previous work). The inclusion of fourth-order effects in a Landau-Zener-resummed kernel is shown to improve both the dephasing rate and the obedience of detailed balance over simpler prescriptions like the non-interacting blip approximation, showing a relatively quick convergence on the exact answer. The results suggest that including higher-order contributions to the memory kernel of a generalized master equation and performing an appropriate resummation can provide a numerically-exact solution to system-bath dynamics for a general spectral density, opening the way to a new class of methods for treating system-bath dynamics.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

Estimation of stochastic volatility with long memory for index prices of FTSE Bursa Malaysia KLCI

In recent years, modeling in long memory properties or fractionally integrated processes in stochastic volatility has been applied in the financial time series. A time series with structural breaks can generate a strong persistence in the autocorrelation function, which is an observed behaviour of a long memory process. This paper considers the structural break of data in order to determine true long memory time series data. Unlike usual short memory models for log volatility, the fractional Ornstein-Uhlenbeck process is neither a Markovian process nor can it be easily transformed into a Markovian process. This makes the likelihood evaluation and parameter estimation for the long memory stochastic volatility (LMSV) model challenging tasks. The drift and volatility parameters of the fractional Ornstein-Unlenbeck model are estimated separately using the least square estimator (lse) and quadratic generalized variations (qgv) method respectively. Finally, the empirical distribution of unobserved volatility is estimated using the particle filtering with sequential important sampling-resampling (SIR) method. The mean square error (MSE) between the estimated and empirical volatility indicates that the performance of the model towards the index prices of FTSE Bursa Malaysia KLCI is fairly well.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

Quasi-unipolar pentacene films embedded with fullerene for non-volatile organic transistor memories

Quasi-unipolar non-volatile organic transistor memory (NOTM) can combine the best characteristics of conventional unipolar and ambipolar NOTMs and, as a result, exhibit improved device performance. Unipolar NOTMs typically exhibit a large signal ratio between the programmed and erased current signals but also require a large voltage to program and erase the memory cells. Meanwhile, an ambipolar NOTM can be programmed and erased at lower voltages, but the resulting signal ratio is small. By embedding a discontinuous n-type fullerene layer within a p-type pentacene film, quasi-unipolar NOTMs are fabricated, of which the signal storage utilizes both electrons and holes while the electrical signal relies on only hole conduction. These devices exhibit superior memory performance relative to both pristine unipolar pentacene devices and ambipolar fullerene/pentacene bilayer devices. The quasi-unipolar NOTM exhibited a larger signal ratio between the programmed and erased states while also reducing the voltage required to program and erase a memory cell. This simple approach should be readily applicable for various combinations of advanced organic semiconductors that have been recently developed and thereby should make a significant impact on organic memory research.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2015

Switching kinetics of SiC resistive memory for harsh environments

Cu/a-SiC/Au resistive memory cells are measured using voltage pulses and exhibit the highest R{sub OFF}/R{sub ON} ratio recorded for any resistive memory. The switching kinetics are investigated and fitted to a numerical model, using thermal conductivity and resistivity properties of the dielectric. The SET mechanism of the Cu/a-SiC/Au memory cells is found to be due to ionic motion without joule heating contributions, whereas the RESET mechanism is found to be due to thermally assisted ionic motion. The conductive filament diameter is extracted to be around 4nm. The high thermal conductivity and resistivity for the Cu/a-SiC/Au memory cells result in slow switching but with high thermal reliability and stability, showing potential for use in harsh environments. Radiation properties of SiC memory cells are investigated. No change was seen in DC sweep or pulsed switching nor in conductive mechanisms, up to 2Mrad(Si) using {sup 60}Co gamma irradiation.

36 MATERIALS SCIENCE↗
DOE OSTI2016

Ambipolar nonvolatile memory based on a quantum-dot transistor with a nanoscale floating gate

Using only solution processing methods, we developed ambipolar quantum-dot (QD) transistor floating-gate memory (FGM) that uses Au nanoparticles as a floating gate. Because of the bipolarity of the active channel of PbSe QDs, the memory could easily trap holes or electrons in the floating gate by programming/erasing (P/E) operations, which could shift the threshold voltage both up and down. As a result, the memory exhibited good programmable memory characteristics: a large memory window (ΔV{sub th} ∼ 15 V) and a long retention time (>10{sup 5 }s). The magnitude of ΔV{sub th} depended on both P/E voltages and the bias voltage (V{sub DS}): ΔV{sub th} was a cubic function to V{sub P/E} and linearly depended on V{sub DS}. Therefore, this FGM based on a QD transistor is a promising alternative to its inorganic counterparts owing to its advantages of bipolarity, high mobility, low cost, and large-area production.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2014

Proteome-wide analysis of HIV-specific naive and memory CD4 + T cells in unexposed blood donors

The preexisting HIV-1–specific T cell repertoire must influence both the immunodominance of T cells after infection and immunogenicity of vaccines. We directly compared two methods for measuring the preexisting CD4 + T cell repertoire in healthy HIV-1–negative volunteers, the HLA-peptide tetramer enrichment and T cell library technique, and show high concordance (r = 0.989). Using the library technique, we examined whether naive, central memory, and/or effector memory CD4 + T cells specific for overlapping peptides spanning the entire HIV-1 proteome were detectable in 10 HLA diverse, HIV-1–unexposed, seronegative donors. HIV-1–specific cells were detected in all donors at a mean of 55 cells/million naive cells and 38.9 and 34.1 cells/million in central and effector memory subsets. Remarkably, peptide mapping showed most epitopes recognized by naive (88%) and memory (56%) CD4 + T cells had been previously reported in natural HIV-1 infection. Furthermore, 83% of epitopes identified in preexisting memory subsets shared epitope length matches (8–12 amino acids) with human microbiome proteins, suggestive of a possible cross-reactive mechanism. These results underline the power of a proteome-wide analysis of peptide recognition by human T cells for the identification of dominant antigens and provide a baseline for optimizing HIV-1–specific helper cell responses by vaccination.

59 BASIC BIOLOGICAL SCIENCES↗