Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory systems”

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 1,729 records · Page 96

Merlin - Massively parallel heterogeneous computing

Hardware and software for Merlin, a new kind of massively parallel computing system, are described. Eight computers are linked as a 300-MIPS prototype to develop system software for a larger Merlin network with 16 to 64 nodes, totaling 600 to 3000 MIPS. These working prototypes help refine a mapped reflective memory technique that offers a new, very general way of linking many types of computer to form supercomputers. Processors share data selectively and rapidly on a word-by-word basis. Fast firmware virtual circuits are reconfigured to match topological needs of individual application programs. Merlin's low-latency memory-sharing interfaces solve many problems in the design of high-performance computing systems. The Merlin prototypes are intended to run parallel programs for scientific applications and to determine hardware and software needs for a future Teraflops Merlin network.

Wittie, Larry↗

Mixed Delay/Nondelay Embeddings Based Neuromorphic Computing with Patterned Nanomagnet Arrays

Patterned nanomagnet arrays (PNAs) have been shown to exhibit a strong geometrically frustrated dipole interaction. Some PNAs have also shown emergent domain wall dynamics. Previous works have demonstrated methods to physically probe these magnetization dynamics of PNAs to realize neuromorphic reservoir systems that exhibit chaotic dynamical behavior and high-dimensional nonlinearity. These PNA reservoir systems from prior works leverage echo state properties and linear/nonlinear short-term memory of component reservoir nodes to map and preserve the dynamical information of the input time-series data into nondelay spatial embeddings. Such mappings enable these PNA reservoir systems to imitate and predict/forecast the input time series data. However, these prior PNA reservoir systems are based solely on the nondelay spatial embeddings obtained at component reservoir nodes. As a result, they require a massive number of component reservoir nodes, or a very large spatial embedding (i.e., high-dimensional spatial embedding) per reservoir node, or both, to achieve acceptable imitation and prediction accuracy. These requirements reduce the practical feasibility of such PNA reservoir systems. To address this shortcoming, we present a mixed delay/nondelay embeddings-based PNA reservoir system. Our system uses a single PNA reservoir node with the ability to obtain a mixture of delay/nondelay embeddings of the dynamical information of the time-series data applied at the input of a single PNA reservoir node. Our analysis shows that when these mixed delay/nondelay embeddings are used to train a perceptron at the output layer, our reservoir system outperforms existing PNA-based reservoir systems for the imitation of NARMA 2, NARMA 5, NARMA 7, and NARMA 10 time series data, and for the short-term and long-term prediction of the Mackey Glass time series data.

Ti, Changpeng↗

Fabrication of vertically aligned ferroelectric polyvinylidene fluoride mesoscale rod arrays

Here, we have fabricated vertically aligned ferroelectric PVDF mesoscale rod arrays comprising and phases using a 200 nm diameter anodized aluminum oxide (AAO) as the porous template. We could synthesize the ferroelectric phase in mesoscale rod forms by combining the well-established recipe for crystallizing the phase using dimethyl sulfoxide (DMSO) at low temperature and template-guided infiltration processing for the rods using AAO. We also measured the dimensions of the PVDF rods by scanning electron microscopy and identified the polymorph phases by X-ray diffraction and Fourier transform infrared spectroscopy. The length of the rods varied from 3.82 m to 1.09 m and the diameter from 232 nm to 287 nm when the volume ratio between DMSO and acetone changed from 5 : 5 to 10 : 0. We obtained well-defined piezoresponse hysteresis loops for all rods with remnant piezoresponse ranging from 2.12 pm/V to 5.04 pm/V and coercive voltage ranging from 2.29 V to 2.71 V using piezoresponse force microscopy. These results serve as a processing platform for flexible electronic devices that need high capacitance and piezoelectric functionalities such as flexible memory devices or body energy harvesting devices for intelligent systems. (c) 2013 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 130: 3842-3848, 2013

36 MATERIALS SCIENCE↗

Drought legacies mediated by trait trade‐offs in soil microbiomes

Abstract Soil microbiomes play a key role in driving biogeochemical cycles of the Earth system. As drought frequency and intensity increase due to climate change, soil microbes and the processes they control will be impacted. Even after a drought ends, microbiomes and other systems take time to recover and may display a memory of previous climate conditions. Still, the mechanisms involved in these legacy effects remain unclear, making it difficult to predict climate and biogeochemical rates in the future. Here, we used a trait‐based microbiome model (DEMENTpy) to implement trade‐off‐mediated mechanisms that may lead to drought legacy effects on litter decomposition. Trade‐offs were assumed to follow the Y‐A‐S framework that defines three primary life‐history strategies of microorganisms: high growth Yield, resource Acquisition, and Stress tolerance. We represented cellular trade‐offs between osmolytes required for drought tolerance and investment in enzymes involved in litter decomposition. Simulations were run under varying levels of drought severity and dispersal. With high levels of dispersal, no legacy effects were predicted by DEMENTpy following drought. With limited dispersal, severe drought resulted in a persistent legacy of altered community‐level traits and reduced litter decomposition. Moderate drought resulted in a transient legacy that disappeared after two years, consistent with recent empirical observations in Southern California ecosystems. These results imply that greater movement along the trade‐off between enzyme investment and osmolyte production resulted in stronger legacy effects. More generally, factors that shift the position of a microbiome in YAS space may alter the legacy outcome following drought. Our trait‐based modeling study motivates additional empirical measurements to quantify YAS traits and trade‐offs that are needed to make accurate predictions of soil microbiome resilience and functioning. Also, our study illustrates an emerging approach for representing trait trade‐offs in microbiomes and vegetation that dictate ecosystem responses to drought and other environmental perturbations.

54 ENVIRONMENTAL SCIENCES↗

Asynchronous Two-Level Checkpointing Scheme for Large-Scale Adjoints in the Spectral-Element Solver Nek5000

Adjoints are an important computational tool for large-scale sensitivity evaluation, uncertainty quantification, and derivative-based optimization. An essential component of their performance is the storage/recomputation balance in which efficient checkpointing methods play a key role. We introduce a novel asynchronous two-level adjoint checkpointing scheme for multistep numerical time discretizations targeted at large-scale numerical simulations. The checkpointing scheme combines bandwidth-limited disk checkpointing and binomial memory checkpointing. Based on assumptions about the target petascale systems, which we later demonstrate to be realistic on the IBM Blue Gene/Q system Mira, we create a model of the expected performance of our checkpointing approach and validate it using the highly scalable Navier-Stokes spectralelement solver Nek5000 on small to moderate subsystems of the Mira supercomputer. In turn, this allows us to predict optimal algorithmic choices when using all of Mira. We also demonstrate that two-level checkpointing is significantly superior to single-level checkpointing when adjoining a large number of time integration steps. To our knowledge, this is the first time two-level checkpointing had been designed, implemented, tuned, and demonstrated on fluid dynamics codes at large scale of 50k+ cores.

Adjoints↗

A Preprocessing Tool for Enhanced Ion Mobility–Mass Spectrometry-Based Omics Workflows

The ability to improve the data quality of ion mobility–mass spectrometry (IM-MS) measurements is of great importance for enabling modular and efficient computational workflows and gaining better qualitative and quantitative insights from complex biological and environmental samples. We developed the PNNL PreProcessor, a standalone and user-friendly software housing various algorithmic implementations to generate new MS-files with enhanced signal quality and in the same instrument format. Different experimental approaches are supported for IM-MS based on Drift-Tube (DT) and Structures for Lossless Ion Manipulations (SLIM), including liquid chromatography (LC) and infusion analyses. The algorithms extend the dynamic range of the detection system, while reducing file sizes for faster and memory-efficient downstream processing. Specifically, multidimensional smoothing improves peak shapes of poorly defined low-abundance signals, and saturation repair reconstructs the intensity profile of high-abundance peaks from various analyte types. Further, other functionalities are data compression and interpolation, IM demultiplexing, noise filtering by low intensity threshold and spike removal, and exporting of acquisition metadata. Several advantages of the tool are illustrated, including an increase of 19.4% in lipid annotations and a two-times faster processing of LC-DT IM-MS data-independent acquisition spectra from a complex lipid extract of a standard human plasma sample. The software is freely available at https://omics.pnl.gov/software/pnnl-preprocessor.

59 BASIC BIOLOGICAL SCIENCES↗

Enhanced read resolution in reconfigurable memristive synapses for Spiking Neural Networks

Abstract The synapse is a key element circuit in any memristor-based neuromorphic computing system. A memristor is a two-terminal analog memory device. Memristive synapses suffer from various challenges including high voltage, SET or RESET failure, and READ margin issues that can degrade the distinguishability of stored weights. Enhancing READ resolution is very important to improving the reliability of memristive synapses. Usually, the READ resolution is very small for a memristive synapse with a 4-bit data precision. This work considers a step-by-step analysis to enhance the READ current resolution or the read current difference between two resistance levels for a current-controlled memristor-based synapse. An empirical model is used to characterize the $${\hbox {HfO}}_{2}$$ HfO 2 based memristive device. $$1\textrm{st}$$ 1 st and $$2\textrm{nd}$$ 2 nd stage device of our proposed synapse design can be scaled to enhance the READ current margin up to $$\sim$$ ∼ 4.3 $$\times$$ × and $$\sim$$ ∼ 21%, respectively. Moreover, READ current resolution can be enhanced with run-time adaptation techniques such as READ voltage scaling and body biasing. The READ voltage scaling and body biasing can improve the READ current resolution by about 46% and 15%, respectively. TENNLab’s neuromorphic computing framework is leveraged to evaluate the effect of READ current resolution on classification, control, and reservoir computing applications. Higher READ current resolution shows better accuracy than lower resolution even when facing different levels of read noise.

97 MATHEMATICS AND COMPUTING↗

Slowing and stopping light with an optomechanical crystal array

The ability to coherently store and retrieve optical information in a rapidly tunable manner is an important ingredient for all-optical information processing. In the classical domain, this optical buffering is necessary to manage information flow in complex networks. In quantum information processing, such a system can also serve as a long-term memory capable of storing the full quantum information contained in an optical pulse. Here we suggest a novel approach to light storage involving an optical waveguide coupled to an optomechanical crystal array, where light in the waveguide can be dynamically and reversibly mapped into long-lived mechanical vibrations in the array. This technique enables large bandwidths and long storage and delay times in a compact, on-chip platform.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

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 ↗

Efficient Streaming Dynamic Mode Decomposition

We propose a reformulation of the streaming dynamic mode decomposition method that requires maintaining a single orthonormal basis, thereby reducing computational redundancy. The proposed efficient streaming dynamic mode decomposition method results in a constant-factor reduction in computational complexity and memory storage requirements. Numerical experiments on representative canonical dynamical systems show that the enhanced computational efficiency does not compromise the accuracy of the proposed method.

97 MATHEMATICS AND COMPUTING↗

Machine Learning Toolkit for Extreme Scale

Support Vector Machines (SVM) is a popular machine learning technique, which has been applied to a wide range of domains such as science, finance, and social networks for supervised learning. MaTEx undertakes the challenge of designing a scalable parallel SVM training algorithm for large scale systems, which includes commodity multi-core machines, tightly connected supercomputers and cloud computing systems. Several techniques are proposed for improved speed and memory space usage including adaptive and aggressive elimination of samples for faster convergence , and sparse format representation of data samples. Several heuristics for earliest possible to lazy elimination of non-contributing samples are considered in MaTEx. In many cases, where an early sample elimination might result in a false positive, low overhead mechanisms for reconstruction of key data structures are proposed. The proposed algorithm and heuristics are implemented and evaluated on various publicly available datasets

Source record↗

Performance Characteristics of HYDRA - a Multi-Physics simulation code from Lawrence Livermore National Laboratory

HYDRA is used to simulate a variety of experiments carried out at the National Ignition Facility (NIF) [4] and other high energy density physics facilities. HYDRA has packages to simulate radiation transfer, atomic physics, hydrodynamics, laser propagation, and a number of other physics effects. HYDRA has over one million lines of code and includes both MPI and thread-level (OpenMP and pthreads) parallelism. This paper measures the performance characteristics of HYDRA using hardware counters on an IBM BlueGene/Q system. We report key ratios such as bytes/instruction and memory bandwidth for several different physics packages. The total number of bytes read and written per time step is also reported. We show that none of the packages which use significant time are memory bandwidth limited on a Blue Gene/Q. HYDRA currently issues very few SIMD instructions. The pressure on memory bandwidth will increase if high levels of SIMD instructions can be achieved.

97 MATHEMATICS AND COMPUTING↗

AEP Ohio gridSMART Demonstration Project Real-Time Pricing Demonstration Analysis

This report contributes initial findings from an analysis of significant aspects of the gridSMART® Real-Time Pricing (RTP) – Double Auction demonstration project. Over the course of four years, Pacific Northwest National Laboratory (PNNL) worked with American Electric Power (AEP), Ohio and Battelle Memorial Institute to design, build, and operate an innovative system to engage residential consumers and their end-use resources in a participatory approach to electric system operations, an incentive-based approach that has the promise of providing greater efficiency under normal operating conditions and greater flexibility to react under situations of system stress. The material contained in this report supplements the findings documented by AEP Ohio in the main body of the gridSMART report. It delves into three main areas: impacts on system operations, impacts on households, and observations about the sensitivity of load to price changes.

Widergren, Steven E.↗

Summer Proceedings 2016: The Center for Computing Research at Sandia National Laboratories

Solving sparse linear systems from the discretization of elliptic partial differential equations (PDEs) is an important building block in many engineering applications. Sparse direct solvers can solve general linear systems, but are usually slower and use much more memory than effective iterative solvers. To overcome these two disadvantages, a hierarchical solver (LoRaSp) based on H2-matrices was introduced in [22]. Here, we have developed a parallel version of the algorithm in LoRaSp to solve large sparse matrices on distributed memory machines. On a single processor, the factorization time of our parallel solver scales almost linearly with the problem size for three-dimensional problems, as opposed to the quadratic scalability of many existing sparse direct solvers. Moreover, our solver leads to almost constant numbers of iterations, when used as a preconditioner for Poisson problems. On more than one processor, our algorithm has significant speedups compared to sequential runs. With this parallel algorithm, we are able to solve large problems much faster than many existing packages as demonstrated by the numerical experiments.

97 MATHEMATICS AND COMPUTING↗

Lightweight storage and overlay networks for fault tolerance.

The next generation of capability-class, massively parallel processing (MPP) systems is expected to have hundreds of thousands to millions of processors, In such environments, it is critical to have fault-tolerance mechanisms, including checkpoint/restart, that scale with the size of applications and the percentage of the system on which the applications execute. For application-driven, periodic checkpoint operations, the state-of-the-art does not provide a scalable solution. For example, on today's massive-scale systems that execute applications which consume most of the memory of the employed compute nodes, checkpoint operations generate I/O that consumes nearly 80% of the total I/O usage. Motivated by this observation, this project aims to improve I/O performance for application-directed checkpoints through the use of lightweight storage architectures and overlay networks. Lightweight storage provide direct access to underlying storage devices. Overlay networks provide caching and processing capabilities in the compute-node fabric. The combination has potential to signifcantly reduce I/O overhead for large-scale applications. This report describes our combined efforts to model and understand overheads for application-directed checkpoints, as well as implementation and performance analysis of a checkpoint service that uses available compute nodes as a network cache for checkpoint operations.

99 GENERAL AND MISCELLANEOUS↗

Between Harmonic Crystal and Glass: Solids with Dimpled Potential-Energy Surfaces Having Multiple Local Energy Minima

Solids with dimpled potential-energy surfaces are ubiquitous in nature and, typically, exhibit structural (elastic or phonon) instabilities. Dimpled potentials are not harmonic; thus, the conventional quasiharmonic approximation at finite temperatures fails to describe anharmonic vibrations in such solids. At sufficiently high temperatures, their crystal structure is stabilized by entropy; in this phase, a diffraction pattern of a periodic crystal is combined with vibrational properties of a phonon glass. As temperature is lowered, the solid undergoes a symmetry-breaking transition and transforms into a lower-symmetry phase with lower lattice entropy. Here, we identify specific features in the potential-energy surface that lead to such polymorphic behavior; we establish reliable estimates for the relative energies and temperatures associated with the anharmonic vibrations and the solid–solid symmetry-breaking phase transitions. We show that computational phonon methods can be applied to address anharmonic vibrations in a polymorphic solid at fixed temperature. To illustrate the ubiquity of this class of materials, we present a range of examples (elemental metals, a shape-memory alloy, and a layered charge-density-wave system); we show that our theoretical predictions compare well with known experimental data.

36 MATERIALS SCIENCE↗