Search NASA⌕ Search

SEARCH · Search NASA

Results for “memory efficiency”

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 1,801 records · Page 100

Lost in space: Onboard star identification using CCD star tracker data without an a priori attitude

There are many algorithms in use today which determine spacecraft attitude by identifying stars in the field of view of a star tracker. Some methods, which date from the early 1960's, compare the angular separation between observed stars with a small catalog. In the last 10 years, several methods have been developed which speed up the process and reduce the amount of memory needed, a key element to onboard attitude determination. However, each of these methods require some a priori knowledge of the spacecraft attitude. Although the Sun and magnetic field generally provide the necessary coarse attitude information, there are occasions when a spacecraft could get lost when it is not prudent to wait for sunlight. Also, the possibility of efficient attitude determination using only the highly accurate CCD star tracker could lead to fully autonomous spacecraft attitude determination. The need for redundant coarse sensors could thus be eliminated at substantial cost reduction. Some groups have extended their algorithms to implement a computation intense full sky scan. Some require large data bases. Both storage and speed are concerns for autonomous onboard systems. Neural network technology is even being explored by some as a possible solution, but because of the limited number of patterns that can be stored and large overhead, nothing concrete has resulted from these efforts. This paper presents an algorithm which, by descretizing the sky and filtering by visual magnitude of the brightness observed star, speeds up the lost in space star identification process while reducing the amount of necessary onboard computer storage compared to existing techniques.

Ketchum, Eleanor A.↗

Implementation of a Message Passing Interface into a Cloud-Resolving Model for Massively Parallel Computing

The capability for massively parallel programming (MPP) using a message passing interface (MPI) has been implemented into a three-dimensional version of the Goddard Cumulus Ensemble (GCE) model. The design for the MPP with MPI uses the concept of maintaining similar code structure between the whole domain as well as the portions after decomposition. Hence the model follows the same integration for single and multiple tasks (CPUs). Also, it provides for minimal changes to the original code, so it is easily modified and/or managed by the model developers and users who have little knowledge of MPP. The entire model domain could be sliced into one- or two-dimensional decomposition with a halo regime, which is overlaid on partial domains. The halo regime requires that no data be fetched across tasks during the computational stage, but it must be updated before the next computational stage through data exchange via MPI. For reproducible purposes, transposing data among tasks is required for spectral transform (Fast Fourier Transform, FFT), which is used in the anelastic version of the model for solving the pressure equation. The performance of the MPI-implemented codes (i.e., the compressible and anelastic versions) was tested on three different computing platforms. The major results are: 1) both versions have speedups of about 99% up to 256 tasks but not for 512 tasks; 2) the anelastic version has better speedup and efficiency because it requires more computations than that of the compressible version; 3) equal or approximately-equal numbers of slices between the x- and y- directions provide the fastest integration due to fewer data exchanges; and 4) one-dimensional slices in the x-direction result in the slowest integration due to the need for more memory relocation for computation.

Juang, Hann-Ming Henry↗

Efficient Runtime Verification of First-Order Temporal Properties

Runtime verification allows monitoring the execution of a system against a temporal property, raising an alarm if the property is violated. In this paper we present a theory and system for runtime verification of a first-order past time linear temporal logic. The first-order nature of the logic allows a monitor to reason about events with data elements. While runtime verification of propositional temporal logic requires only a fixed amount of memory, the first-order variant has to deal with a number of data values potentially growing unbounded in the length of the execution trace. This requires special compactness considerations in order to allow checking very long executions. In previous work we presented an efficient use of BDDs for such first-order runtime verification, implemented in the tool DEJAVU. We first summarize this previous work. Subsequently, we look at the new problem of dynamically identifying when data observed in the past are no longer needed, allowing to reclaim the data elements used to represent them. We also study the problem of adding relations over data values. Finally, we present parts of the implementation, including a new concept of user defined property macros.

Peled, Doron↗

On the Feasibility of 1T Ferroelectric FET Memory Array

To fully exploit the ferroelectric field effect transistor (FeFET) as compact embedded nonvolatile memory for various computing and storage applications, it is desirable to use a single FeFET (1T) as a unit cell and arrange the cells into an array. However, many write mechanisms for an 1T FeFET array reported in the literature are yet to be validated experimentally. In this work, we performed a comprehensive experimental characterization on the write operations in an 1T- NOR and 1T- AND array using n-channel bulk FeFETs. We discovered that: 1) the source/drain contact can only supply minority carriers (i.e., electrons) to the channel for polarization screening during the low- V TH state programming; 2) the body contact can not only supply majority carriers (i.e., holes) for efficient high- V TH state write, but also depletion charge for low- V TH state programming, though with lower efficiency; and 3) during the low/high- V TH programming, only the path that can supply negative/positive screening charges, respectively, need to respond, which necessitates the application of proper write biases on the corresponding terminal. Based on the understanding of these write mechanisms, we show the importance of localized body contact or column-wise body contact for the successful high- V TH state programming. We also show that the previously proposed C- AND write scheme fails to program the FeFET to the low- V TH state for our devices. Finally, we propose several write schemes for both 1T- AND and 1T- NOR arrays for various scenarios providing insights for choosing the appropriate write scheme, which will facilitate the adoption of 1T FeFET memory arrays for emerging applications.

FeFET↗

Belle II Conditions Database

The Belle II experiment at KEK is preparing for taking first collision data in early 2018. For the success of the experiment it is essential to have information about varying conditions available to systems worldwide in a fast and efficient manner that is straightforward for both the user and maintainer. The Belle II Conditions Database was designed to make maintenance as easy as possible. To this end, a HTTP REST service was developed with industry-standard tools such as Swagger for the API interface development, Payara for the Java EE application server, and the Hazelcast in-memory data grid for support of scalable caching as well as transparent distribution of the service across multiple sites. On the client side, the online and offline software has to be able to obtain conditions data from the Belle II Conditions Database in a robust and reliable way under very different situations. As such the client side interface to the Belle II Conditions Database has been designed with a variety of access mechanisms which allow the software to be used with and without an internet connection. Different methods to access the payload information are implemented to allow for a high level of customization per site and to simplify testing of new payloads locally. Changes to the conditions data are usually handled transparently but users can actively check whether an object has changed or register callback functions to be called whenever a conditions data object is updated. In addition a command line user interface has been developed to simplify inspection and modification of the database contents.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Efficient compression of quantum information

We propose a scheme for an exact efficient transformation of a tensor product state of many identically prepared qubits into a state of a logarithmically small number of qubits. Using a quadratic number of elementary quantum gates we transform N identically prepared qubits into a state, which is nontrivial only on the first [log{sub 2}(N+1)] qubits. This procedure might be useful for quantum memories, as only a small portion of the original qubits has to be stored. Another possible application is in communicating a direction encoded in a set of quantum states, as the compressed state provides a high-effective method for such an encoding.

74 ATOMIC AND MOLECULAR PHYSICS↗

SU-F-T-370: A Fast Monte Carlo Dose Engine for Gamma Knife

Purpose: To develop a fast Monte Carlo dose calculation algorithm for Gamma Knife. Methods: To make the simulation more efficient, we implemented the track repeating technique on GPU. We first use EGSnrc to pre-calculate the photon and secondary electron tracks in water from two mono-energy photons of 60Co. The total photon mean free paths for different materials and energies are obtained from NIST. During simulation, each entire photon track was first loaded to shared memory for each block, the incident original photon was then splitted to Nthread sub-photons, each thread transport one sub-photon, the Russian roulette technique was applied for scattered and bremsstrahlung photons. The resultant electrons from photon interactions are simulated by repeating the recorded electron tracks. The electron step length is stretched/shrunk proportionally based on the local density and stopping power ratios of the local material. Energy deposition in a voxel is proportional to the fraction of the equivalent step length in that voxel. To evaluate its accuracy, dose deposition in a 300mm*300mm*300mm water phantom is calculated, and compared to EGSnrc results. Results: Both PDD and OAR showed great agreements (within 0.5%) between our dose engine result and the EGSnrc result. It only takes less than 1 min for every simulation, being reduced up to ∼40 times compared to EGSnrc simulations. Conclusion: We have successfully developed a fast Monte Carlo dose engine for Gamma Knife.

60 APPLIED LIFE SCIENCES↗

Addressing Global Data Dependencies in Heterogeneous Asynchronous Runtime Systems on GPUs

Large-scale parallel applications with complex global data dependencies beyond those of reductions pose significant scalability challenges in an asynchronous runtime system. Internodal challenges include identifying the all-to-all communication of data dependencies among the nodes. Intranodal challenges include gathering together these data dependencies into usable data objects while avoiding data duplication. This paper addresses these challenges within the context of a large-scale, industrial coal boiler simulation using the Uintah asynchronous many-task runtime system on GPU architectures. We show significant reduction in time spent analyzing data dependencies through refinements in our dependency search algorithm. Multiple task graphs are used to eliminate subsequent analysis when task graphs change in predictable and repeatable ways. Using a combined data store and task scheduler redesign reduces data dependency duplication ensuring that problems fit within host and GPU memory. Furthermore, these modifications did not require any changes to application code or sweeping changes to the Uintah runtime system. We report results running on the DOE Titan system on 119K CPU cores and 7.5K GPUs simultaneously. Our solutions can be generalized to other task dependency problems with global dependencies among thousands of nodes which must be processed efficiently at large scale.

97 MATHEMATICS AND COMPUTING↗

Characterizing and Mitigating Work Time Inflation in Task Parallel Programs

Task parallelism raises the level of abstraction in shared memory parallel programming to simplify the development of complex applications. However, task parallel applications can exhibit poor performance due to thread idleness, scheduling overheads, and work time inflation – additional time spent by threads in a multithreaded computation beyond the time required to perform the same work in a sequential computation. We identify the contributions of each factor to lost efficiency in various task parallel OpenMP applications and diagnose the causes of work time inflation in those applications. Increased data access latency can cause significant work time inflation in NUMA systems. Our locality framework for task parallel OpenMP programs mitigates this cause of work time inflation. Our extensions to the Qthreads library demonstrate that locality-aware scheduling can improve performance up to 3X compared to the Intel OpenMP task scheduler.

Olivier, Stephen L.↗

Machine Learning Assisted Safety Modeling and Analysis of Advanced Reactors

With the advances in computational power and numerical methods, analysts can now rely on first-principle simulations to predict ultra-fine details in a variety of applications. Advances in machine learning (ML) have produced algorithms that can now learn high-level abstractions via hierarchical models. This project aims to leverage advances in ML techniques and the available high-resolution simulation data to develop a novel modeling and simulation (M\&S) methodology for reactor safety analysis. While application-agnostic ML techniques are available, complex physics constraints need to be incorporated into ML techniques to build ML-based closures for computationally efficient predictive simulations. This project intends to develop a physics-guided data-driven multi-scale methodology for M\&S of advanced reactors. The project focuses on thermal fluid (T/F) phenomena, which play major roles in advanced reactor safety. Specifically, we propose a data-driven coarse-mesh turbulence model based on local flow features for the transient analysis of thermal mixing and stratification in a sodium-cooled fast reactor (SFR). The model has a coarse-mesh setup to ensure computational efficiency, while it is trained by fine-mesh computational fluid dynamics (CFD) data with Reynolds-averaged Navier-Stokes (RANS) turbulence model to ensure accuracy. Three different neural networks are developed and tested for loss-of-flow transients in the hot pool of SFR, i.e. the densely connected convolutional neural network (DCNN), long-short-term-memory network based on proper orthogonal decomposition (POD-LSTM), and the DCNN informed by LSTM (DCNN-LSTM). The performances of these three neural networks are evaluated based on baseline models. The DCNN-LSTM model has been chosen for further hyperparameter optimization. Furthermore, based on a simplified two-dimensional case, uncertainty quantification (UQ) of the developed ML-based closure are investigated with three methods, i.e. Monte Carlo dropout, deep ensemble, and Bayesian neural network. The developed ML-based turbulent viscosity closure relation based on deep ensemble is then integrated into the system analysis module SAM and serves as a term in the conservation equations. Such a SAM-ML based procedure guarantees that the obtained results are consistent with the physical constraints of the thermal-fluid system. The SAM-ML simulation on the same loss-of-flow transient showed comparable accuracy with the CFD simulation but with a much coarser mesh setup. Last but not least, the ML-based closure improvement with the support of higher-fidelity data from large eddy simulation (LES) is discussed. As a first step towards this direction, a baseline LES simulation is performed to obtain comparable data with RANS results. Based on the early results, future investigation on further improving the ML-based closure is discussed. We believe the developed approach that combines scientific machine learning with nuclear system analysis code can benefit the advanced reactor community as more accurate safety analyses will better characterize reactor safety margins and reduce licensing efforts.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

A 3D agglomeration multigrid solver for the Reynolds-averaged Navier-Stokes equations on unstructured meshes

An agglomeration multigrid strategy is developed and implemented for the solution of three-dimensional steady viscous flows. The method enables convergence acceleration with minimal additional memory overheads, and is completely automated, in that it can deal with grids of arbitrary construction. The multigrid technique is validated by comparing the delivered convergence rates with those obtained by a previously developed overset-mesh multigrid approach, and by demonstrating grid independent convergence rates for aerodynamic problems on very large grids. Prospects for further increases in multigrid efficiency for high-Reynolds number viscous flows on highly stretched meshes are discussed.

Marvriplis, D. J.↗

Uncertainty quantification of machine learning models to improve streamflow prediction under changing climate and environmental conditions

Machine learning (ML) models, and Long Short-Term Memory (LSTM) networks in particular, have demonstrated remarkable performance in streamflow prediction and are increasingly being used by the hydrological research community. However, most of these applications do not include uncertainty quantification (UQ). ML models are data driven and can suffer from large extrapolation errors when applied to changing climate/environmental conditions. UQ is required to quantify the influence of data noises on model predictions and avoid overconfident projections in extrapolation. In this work, we integrate a novel UQ method, called PI3NN, with LSTM networks for streamflow prediction. PI3NN calculates Prediction Intervals by training 3 Neural Networks. It can precisely quantify the predictive uncertainty caused by the data noise and identify out-of-distribution (OOD) data in a non-stationary condition to avoid overconfident predictions. We apply the PI3NN-LSTM method in the snow-dominant East River Watershed in the western US and in the rain-driven Walker Branch Watershed in the southeastern US. Results indicate that for the prediction data which have similar features as the training data, PI3NN precisely quantifies the predictive uncertainty with the desired confidence level; and for the OOD data where the LSTM network fails to make accurate predictions, PI3NN produces a reasonably large uncertainty indicating that the results are not trustworthy and should avoid overconfidence. PI3NN is computationally efficient, robust in performance, and generalizable to various network structures and data with no distributional assumptions. It can be broadly applied in ML-based hydrological simulations for credible prediction.

54 ENVIRONMENTAL SCIENCES↗

A common intermediate for N2 formation in enzymes and zeolites: side-on Cu-nitrosyl complexes

Understanding the mechanisms of catalytic processes requires the identification of reaction centers and key intermediates, both of which are often achieved by the use of spectroscopic characterization tools. Due to the heterogeneity of active centers in heterogeneous catalysts, it is frequently difficult to identify the specific sites that are responsible for the overall activity. Furthermore, the simultaneous presence of a large number of surface species on the catalyst surface often poses a great challenge for the unambiguous determination of the relevant species in the reaction mechanism. In contrast, enzymes possess catalytically active centers with precisely defined coordination environments that are only able to accommodate intermediates relevant to the specific catalytic process. Here we show that side-on Cu+-NO+ complexes characterized by high magnetic field solid state magic angle spinning nuclear magnetic resonance (NMR) and Fourier transform infrared (FTIR) spectroscopies are the key intermediates in the selective catalytic reduction of NO over Cu-SSZ-13 zeolite catalysts. Analogous intermediates have been observed and characterized in nitrite reductase enzymes, and shown to be the critical intermediates in the formation of N2 for anaerobic ammonium oxidation reactions.[1] The identification of this key reaction intermediate, combined with the results of our prior kinetic studies, allows us to propose a new reaction mechanism for the selective catalytic reduction of NO with NH3 under oxygen-rich environments over Cu-SSZ-13 zeolites, a key reaction in automotive emission control. The authors acknowledge the US Department of Energy (DOE), Office of Energy Efficiency and Renewable Energy/Vehicle Technologies Program for the support of this work. The research described in this paper was performed at the Environmental Molecular Sciences Laboratory (EMSL), a national scientific user facility sponsored by the DOE’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory (PNNL). PNNL is operated for the US DOE by Battelle Memorial Institute.

Environmental Molecular Sciences Laboratory↗

Developmental regulation of RNA processing by Rbfox proteins

The Rbfox genes encode an ancient family of sequence-specific RNA binding proteins (RBPs) that are critical developmental regulators in multiple tissues including skeletal muscle, cardiac muscle, and brain. The hallmark of Rbfox proteins is a single high-affinity RRM domain, highly conserved from insects to humans, that binds preferentially to UGCAUG motifs at diverse regulatory sites in pre-mRNA introns, mRNA 3’UTRs, and pre-miRNAs hairpin structures. These versatile regulatory circuits operate on Rbfox pre-mRNA and mRNA to ensure proper expression of Rbfox1 protein isoforms, which then act on the broader transcriptome to regulate alternative splicing networks, mRNA stability and translation, and microRNA processing. Complex Rbfox expression is encoded in large genes encompassing multiple promoters and alternative splicing options that govern spatiotemporal expression of structurally distinct and tissue-specific protein isoforms with different classes of RNA targets. Nuclear Rbfox1 is a candidate master regulator that binds intronic UGCAUG elements to impact splicing efficiency of target alternative exons, many in transcripts for other splicing regulators. Tissue-specificity of Rbfox-mediated alternative splicing is executed by combinatorial regulation through the integrated activity of Rbfox proteins and synergistic or antagonistic splicing factors. Studies in animal models show that Rbfox1-related genes are critical for diverse developmental processes including germ cell differentiation and memory in Drosophila, neuronal migration and function in mouse brain, myoblast fusion and skeletal muscle function, and normal heart function. Finally, genetic and biochemical evidence suggest that aberrations in Rbfox-regulated circuitry are risk factors for multiple human disorders, especially neurodevelopmental disorders including epilepsy and autism, and cardiac hypertrophy.

59 BASIC BIOLOGICAL SCIENCES↗

Calculating the divided differences of the exponential function by addition and removal of inputs

We introduce a method for calculating the divided differences of the exponential function by means of addition and removal of items from the input list to the function. Our technique exploits a new identity related to divided differences recently derived by F. Zivcovich. We show that upon adding an item to or removing an item from the input list of an already evaluated exponential, the re-evaluation of the divided differences can be done with only O(sn) floating point operations and O(sn) bytes of memory, where [z 0 ,...,z n ] are the inputs and s ∝ max i,j |z i – z j |. We demonstrate our algorithm’s ability to deal with input lists that are orders-of-magnitude longer than the maximal capacities of the current state-of-the-art. Here, we discuss in detail one practical application of our method: the efficient calculation of weights in the off-diagonal series expansion quantum Monte Carlo algorithm.

97 MATHEMATICS AND COMPUTING↗

Contraction of fermionic operator circuits and the simulation of strongly correlated fermions

A fermionic operator circuit is a product of fermionic operators of usually different and partially overlapping support. Further elements of fermionic operator circuits (FOCs) are partial traces and partial projections. The presented framework allows for the introduction of fermionic versions of known qudit operator circuits (QUOC), important for the simulation of strongly correlated d-dimensional systems: the multiscale entanglement renormalization ansaetze (MERA), tree tensor networks (TTN), projected entangled pair states (PEPS), or their infinite-size versions (iPEPS etc.). After the definition of a FOC, we present a method to contract it with the same computation and memory requirements as a corresponding QUOC, for which all fermionic operators are replaced by qudit operators of identical dimension. A given scheme for contracting the QUOC relates to an analogous scheme for the corresponding fermionic circuit, where additional marginal computational costs arise only from reordering of modes for operators occurring in intermediate stages of the contraction. Our result hence generalizes efficient schemes for the simulation of d-dimensional spin systems, as MERA, TTN, or PEPS to the fermionic case.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A technique for real-time data preprocessing

A processing system is presented that implements simultaneously the efficiency of the special-purpose processor and the total applicability of the general-purpose computer - characteristics commonly thought of as being mutually exclusive. The solution adopted is that of specializing the machine by programming the hardware structure, rather than by adding software systems to it. Data are organized in circulating pages which form a plurality of local dynamic memories for each process. Programs are made up of modules, each describing a transient special-purpose machine. Applications to real-time processing of radar signals are referred to.

Schaffner, M. R.↗

On-board ephemeris representation for Topex/Poseidon

The Topex/Poseidon satellite requires real-time on-board knowledge of the satellite and TDRS ephemeris for attitude determination and control and High-Gain Antenna (HGA) pointing. The ephemeris representation concept for the MMS (Multimission Modular Spacecraft) satellites has shown that compressing the predicted ephemeris in a Fourier Power Series (FPS) before uplinking in conjunction with the On-Board Computer (OBC) ephemeris reconstruction algorithms is an efficient technique for ephemeris representation. As an MMS-based satellite, Topex/Poseidon has inherited the Landsat ephemeris representation concept including a daily FPS upload. This paper presents the Topex/Poseidon concept, analysis, and results including the conclusion that the ephemeris representation duration could be extended to 10 days or more and convenient weekly uploading is adopted without an increase in OBC memory requirements.

Salama, Ahmed H.↗

Refine your search to explore more results.