Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory tool”

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 541 records · Page 30

A high-performance parallel algorithm for nonnegative matrix factorization

Non-negative matrix factorization (NMF) is the problem of determining two non-negative low rank factors W and H, for the given input matrix A, such that A ≈ WH. NMF is a useful tool for many applications in different domains such as topic modeling in text mining, background separation in video analysis, and community detection in social networks. Despite its popularity in the data mining community, there is a lack of efficient distributed algorithms to solve the problem for big data sets. We propose a high-performance distributed-memory parallel algorithm that computes the factorization by iteratively solving alternating non-negative least squares (NLS) subproblems for W and H. It maintains the data and factor matrices in memory (distributed across processors), uses MPI for interprocessor communication, and, in the dense case, provably minimizes communication costs (under mild assumptions). As opposed to previous implementations, our algorithm is also flexible: (1) it performs well for both dense and sparse matrices, and (2) it allows the user to choose any one of the multiple algorithms for solving the updates to low rank factors W and H within the alternating iterations. We demonstrate the scalability of our algorithm and compare it with baseline implementations, showing significant performance improvements.

Kannan, Ramakrishnan↗

HPC-NMF: A High-Performance Parallel Algorithm for Nonnegative Matrix Factorization

NMF is a useful tool for many applications in different domains such as topic modeling in text mining, background separation in video analysis, and community detection in social networks. Despite its popularity in the data mining community, there is a lack of efficient distributed algorithms to solve the problem for big data sets. We propose a high-performance distributed-memory parallel algorithm that computes the factorization by iteratively solving alternating non-negative least squares (NLS) subproblems for $\WW$ and $\HH$. It maintains the data and factor matrices in memory (distributed across processors), uses MPI for interprocessor communication, and, in the dense case, provably minimizes communication costs (under mild assumptions). As opposed to previous implementation, our algorithm is also flexible: It performs well for both dense and sparse matrices, and allows the user to choose any one of the multiple algorithms for solving the updates to low rank factors $\WW$ and $\HH$ within the alternating iterations.

Kannan, Ramakrishnan↗

Functional expansion representations of artificial neural networks

In the past few years, significant interest has developed in using artificial neural networks to model and control nonlinear dynamical systems. While there exists many proposed schemes for accomplishing this and a wealth of supporting empirical results, most approaches to date tend to be ad hoc in nature and rely mainly on heuristic justifications. The purpose of this project was to further develop some analytical tools for representing nonlinear discrete-time input-output systems, which when applied to neural networks would give insight on architecture selection, pruning strategies, and learning algorithms. A long term goal is to determine in what sense, if any, a neural network can be used as a universal approximator for nonliner input-output maps with memory (i.e., realized by a dynamical system). This property is well known for the case of static or memoryless input-output maps. The general architecture under consideration in this project was a single-input, single-output recurrent feedforward network.

Gray, W. Steven↗

Development of Metallic Sensory Alloys

Existing nondestructive evaluation (NDE) technologies are inherently limited by the physical response of the structural material being inspected and are therefore not generally effective at the identification of small discontinuities, making the detection of incipient damage extremely difficult. One innovative solution to this problem is to enhance or complement the NDE signature of structural materials to dramatically improve the ability of existing NDE tools to detect damage. To address this need, a multifunctional metallic material has been developed that can be used in structural applications. The material is processed to contain second phase sensory particles that significantly improve the NDE response, enhancing the ability of conventional NDE techniques to detect incipient damage both during and after flight. Ferromagnetic shape-memory alloys (FSMAs) are an ideal material for these sensory particles as they undergo a uniform and repeatable change in both magnetic properties and crystallographic structure (martensitic transformation) when subjected to strain and/or temperature changes which can be detected using conventional NDE techniques. In this study, the use of a ferromagnetic shape memory alloy (FSMA) as the sensory particles was investigated.

Wallace Terryl A.↗

Design and Analysis Tools for Concurrent Blackboard Systems

A blackboard system consists of a set of knowledge sources, a blackboard data structure, and a control strategy used to activate the knowledge sources. The blackboard model of problem solving is best described by Dr. H. Penny Nii of the Stanford University AI Laboratory: "A Blackboard System can be viewed as a collection of intelligent agents who are gathered around a blackboard, looking at pieces of information written on it, thinking about the current state of the solution, and writing their conclusions on the blackboard as they generate them. " The blackboard is a centralized global data structure, often partitioned in a hierarchical manner, used to represent the problem domain. The blackboard is also used to allow inter-knowledge source communication and acts as a shared memory visible to all of the knowledge sources. A knowledge source is a highly specialized, highly independent process that takes inputs from the blackboard data structure, performs a computation, and places the results of the computation in the blackboard data structure. This design allows for an opportunistic control strategy. The opportunistic problem-solving technique allows a knowledge source to contribute towards the solution of the current problem without knowing which of the other knowledge sources will use the information. The use of opportunistic problem-solving allows the data transfers on the blackboard to determine which processes are active at a given time. Designing and developing blackboard systems is a difficult process. The designer is trying to balance several conflicting goals and achieve a high degree of concurrent knowledge source execution while maintaining both knowledge and semantic consistency on the blackboard. Blackboard systems have not attained their apparent potential because there are no established tools or methods to guide in their construction or analyze their performance.

McManus, John W.↗

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↗

AMReX: a framework for block-structured adaptive mesh refinement

AMReX is a C++ software framework that supports the development of block-structured adaptive mesh refinement (AMR) algorithms for solving systems of partial differential equations (PDEs) with complex boundary conditions on current and emerging architectures. AMR reduces the computational cost and memory footprint compared to a uniform mesh while preserving the local descriptions of different physical processes in complex multiphysics algorithms. Current AMReX-based application codes span a number of areas, including atmospheric modeling, astrophysics, combustion, cosmology, fluctuating hydrodynamics, multiphase flows, and particle accelerators. In particular, the AMReX-Astro GitHub repository holds a number of astrophysical modeling tools based on AMReX. The origins of AMReX trace back to the BoxLib software framework. AMReX supports a number of different time-stepping strategies and spatial discretizations. Solution strategies supported by AMReX range from level-by-level approaches (with or without subcycling in time) with multilevel synchronization to full-hierarchy approaches, and any combination thereof. User-defined kernels that operate on patches of data can be written in C++ or Fortran; there is also a Fortran-interface functionality which wraps the core C++ data structures and operations in Fortran wrappers so that an application code based on AMReX can be written entirely in Fortran.

97 MATHEMATICS AND COMPUTING↗

Developing a Hydrological Monitoring and Sub-Seasonal to Seasonal Forecasting System for South and Southeast Asian River Basins

South and Southeast Asia is subject to significant hydrometeorological extremes, including drought. Under rising temperatures, growing populations, and an apparent weakening of the South Asian monsoon in recent decades, concerns regarding drought and its potential impacts on water and food security are on the rise. Reliable sub-seasonal to seasonal (S2S) hydrological forecasts could, in principle, help governments and international organizations to better assess risk and act in the face of an oncoming drought. Here, we leverage recent improvements in S2S meteorological forecasts and the growing power of Earth Observations to provide more accurate monitoring of hydrological states for forecast initialization. Information from both sources is merged in a South and Southeast Asia sub-seasonal to seasonal hydrological forecasting system (SAHFS-S2S), developed collaboratively with the NASA SERVIR program and end-users across the region. This system applies the Noah-MultiParameterization (NoahMP) Land Surface Model (LSM) in the NASA Land Information System (LIS), driven by downscaled meteorological fields from the Global Data Assimilation System (GDAS) and Climate Hazards InfraRed Precipitation products (CHIRP and CHIRPS) to optimize initial conditions. The NASA Goddard Earth Observing System Model - sub-seasonal to seasonal (GEOS-S2S) forecasts, downscaled using the National Center for Atmospheric Research (NCAR) General Analog Regression Downscaling (GARD) tool and quantile mapping, are then applied to drive 5-km resolution hydrological forecasts to a 9-month forecast time horizon. Results show that the skillful predictions of root zone soil moisture can be made one to two months in advance for forecasts initialized in rainy seasons and up to 8 months when initialized in dry seasons. The memory of accurate initial conditions can positively contribute to forecast skills throughout the entire 9-month prediction period in areas with limited precipitation. This SAHFS-S2S has been operationalized at the International Centre for Integrated Mountain Development (ICIMOD) to support drought monitoring and warning needs in the region.

Yifan Zhou↗

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↗

Recent Additions in the Modeling Capabilities of an Open-Source Wave Energy Converter Design Tool: Preprint

WEC-Sim is a midfidelity numerical tool for modeling wave energy conversion devices. The code uses the MATLAB SimMechanics package to solve multibody dynamics and models wave interactions using hydrodynamic coefficients derived from frequency-domain boundary-element methods. This paper presents the new modeling features introduced in the latest release of WEC-Sim. The first feature discussed conversion of the fluid memory kernel to a state-space form. This enhancement offers a substantial computational benefit after the hydrodynamic body-to-body coefficients are introduced and the number of interactions increases exponentially with each additional body. Additional features include the ability to calculate the wave-excitation forces based on the instantaneous incident wave angle, allowing the device to weathervane, as well as import a user-defined wave elevation time series. A review of the hydrodynamic theory for each feature is provided and the successful implementation is verified using test cases.

16 TIDAL AND WAVE POWER↗

Roofline model toolkit: A practical tool for architectural and program analysis

We present preliminary results of the Roofline Toolkit for multicore, many core, and accelerated architectures. This paper focuses on the processor architecture characterization engine, a collection of portable instrumented micro benchmarks implemented with Message Passing Interface (MPI), and OpenMP used to express thread-level parallelism. These benchmarks are specialized to quantify the behavior of different architectural features. Compared to previous work on performance characterization, these microbenchmarks focus on capturing the performance of each level of the memory hierarchy, along with thread-level parallelism, instruction-level parallelism and explicit SIMD parallelism, measured in the context of the compilers and run-time environments. We also measure sustained PCIe throughput with four GPU memory managed mechanisms. By combining results from the architecture characterization with the Roofline model based solely on architectural specifications, this work offers insights for performance prediction of current and future architectures and their software systems. To that end, we instrument three applications and plot their resultant performance on the corresponding Roofline model when run on a Blue Gene/Q architecture.

97 MATHEMATICS AND COMPUTING↗

Development of a SCALE Tool for Continuous-Energy Eigenvalue Sensitivity Coefficient Calculations

Two methods for calculating eigenvalue sensitivity coefficients in continuous-energy Monte Carlo applications were implemented in the KENO code within the SCALE code package. The methods were used to calculate sensitivity coefficients for several criticality safety problems and produced sensitivity coefficients that agreed well with both reference sensitivities and multigroup TSUNAMI-3D sensitivity coefficients. The newly developed CLUTCH method was observed to produce sensitivity coefficients with high figures of merit and low memory requirements, and both continuous-energy sensitivity methods met or exceeded the accuracy of the multigroup TSUNAMI-3D calculations.

CLUTCH↗

Runtime Techniques to Enable a Highly-Scalable Global Address Space Model for Petascale Computing

Over the past decade, the trajectory to the petascale has been built on increased complexity and scale of the underlying parallel architectures. Meanwhile, software developers have struggled to provide tools that maintain the productivity of computational science teams using these new systems. In this regard, Global Address Space (GAS) programming models provide a straightforward and easy to use addressing model, which can lead to improved productivity. However, the scalability of GAS depends directly on the design and implementation of the runtime system on the target petascale distributed-memory architecture. In this paper, we describe the design, implementation, and optimization of the Aggregate Remote Memory Copy Interface (ARMCI) runtime library on the Cray XT5 2.3 PetaFLOPs computer at Oak Ridge National Laboratory. We optimized our implementation with the flow intimation technique that we have introduced in this paper. Our optimized ARMCI implementation improves scalability of both the Global Arrays programming model and a real-world chemistry application—NWChem—from small jobs up through 180,000 cores.

Computer Science↗

Scaling the Earth System Grid to 100Gbps Networks

The SC11 demonstration, titled Scaling the Earth System Grid to 100Gbps Networks, showed the ability to use underlying infrastructure for the movement of climate data over 100Gbps network. Climate change research is one of the critical data intensive sciences, and the amount of data is continuously growing. Climate simulation data is geographically distributed over the world, and it needs to be accessed from many sources for fast and efficient analysis and inter-comparison of simulations. We used a 100Gbps link connecting National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory (LBNL), Argonne National Laboratory (ANL) and Oak Ridge National Laboratory (ORNL). In the demo, the Intergovernmental Panel on Climate Change (IPCC) Fourth Assessment Report (AR4) phase 3 of the Coupled Model Intercomparison Project (CMIP-3) dataset was staged into the memory of computing nodes at ANL and ORNL from NERSC over the 100Gbps network for analysis and visualization. In general, climate simulation data consists of relatively small and large files with irregular file size distribution in each dataset. In this demo, we addressed challenges on data management in terms of high bandwidth networks, usability of existing protocols and middleware tools, and how applications can adapt and benefit from next generation networks.

97 MATHEMATICS AND COMPUTING↗

Fast Quantum Algorithm for Predicting Descriptive Statistics of Stochastic Processes

Stochastic processes are used as a modeling tool in several sub-fields of physics, biology, and finance. Analytic understanding of the long term behavior of such processes is only tractable for very simple types of stochastic processes such as Markovian processes. However, in real world applications more complex stochastic processes often arise. In physics, the complicating factor might be nonlinearities; in biology it might be memory effects; and in finance is might be the non-random intentional behavior of participants in a market. In the absence of analytic insight, one is forced to understand these more complex stochastic processes via numerical simulation techniques. In this paper we present a quantum algorithm for performing such simulations. In particular, we show how a quantum algorithm can predict arbitrary descriptive statistics (moments) of N-step stochastic processes in just O(square root of N) time. That is, the quantum complexity is the square root of the classical complexity for performing such simulations. This is a significant speedup in comparison to the current state of the art.

Williams Colin P.↗

CRCNS22 Learning Rules in the Hippocampus and their Mapping to Neuromorphic Systems (Final Technical Report)

Large scale biologically-realistic computational models are key to investigating the interplay between structure and function in nervous systems, thus paving the way to new clinical methods and neuro-inspired computing solutions. This project focuses on the hippocampus, in particular the CA3-CA1 regions, due to their role in associative learning and memory, pattern separation and completion, and spatial navigation. Investigations into the neuronal organization and learning rule(s) of this circuit can shed light into how declarative memories are formed, stored, recalled and forgotten and inform computational, experimental and clinical neuroscience work. Our project aims at developing a novel data-driven methodology supported by a broad heterogeneous base of neuroscience experimental knowledge and inspired from advances in computer science and engineering. Specifically, this work will benchmark existing and new learning rules within a full-scale spiking neural network simulation of the CA3-CA1 region. The model will be based on an open-source repository, called the Hippocampome, which contains neuronal morphologies, firing patterns, synapse probabilities, and most other required parameters for all known neuron types in the rodent hippocampal formation. The model will be first trained in a supervised fashion for associative memory tasks using backpropagation through time traditionally used in computer science, enhanced with a new technique called the surrogate gradient method. This optimization method will be used to obtain a global loss minimization, but it is not biologically inspired as it assumes the use of data not locally available to the synapses. However, we propose its use as a benchmarking tool, to compare the training performance of local biologically plausible and hardware-mappable learning rules at scale. New rules or combinations will be proposed and tested as needed, based on the obtained results. Progress in this area will also drive the development of novel hardware-mappable algorithms for continual lifelong learning and categorization of new events from few presented examples. This project goes beyond the existing state-of-the-art by looking at large scale realistic neuronal circuits as networks trainable via global optimization methods such as surrogate gradient descent. The objective function of the brain that supports learning is largely unknown, but it is likely that it operates through local learning rules. Studying network trajectories around local minima as proposed in this work represents a useful strategy for understanding whether a network is training by using a specific (set of) learning rule(s). Starting from a completely untrained network is a challenging test since it is difficult to determine how the learning rule affects the trajectory of the network. This interdisciplinary project will help understand what rule governs learning in these regions or if multiple learning rules are involved. The work will develop a robust methodology to measure if the network is converging to the target solution, oscillating around it, or diverging away.

59 BASIC BIOLOGICAL SCIENCES↗

Towards Flange-to-Flange Turbopump Simulations for Liquid Rocket Engines

The primary objective of this research is to support the design of liquid rocket systems for the Advanced Space Transportation System. Since the space launch systems in the near future are likely to rely on liquid rocket engines, increasing the efficiency and reliability of the engine components is an important task. One of the major problems in the liquid rocket engine is to understand fluid dynamics of fuel and oxidizer flows from the fuel tank to plume. Understanding the flow through the entire turbopump geometry through numerical simulation will be of significant value toward design. This will help to improve safety of future space missions. One of the milestones of this effort is to develop, apply and demonstrate the capability and accuracy of 3D CFD methods as efficient design analysis tools on high performance computer platforms. The development of the MPI and MLP versions of the INS3D code is currently underway. The serial version of INS3D code is a multidimensional incompressible Navier-Stokes solver based on overset grid technology. INS3D-MPI is based on the explicit massage-passing interface across processors and is primarily suited for distributed memory systems. INS3D-MLP is based on multi-level parallel method and is suitable for distributed-shared memory systems. For the entire turbopump simulations, moving boundary capability and an efficient time-accurate integration methods are build in the flow solver. To handle the geometric complexity and moving boundary problems, overset grid scheme is incorporated with the solver that new connectivity data will be obtained at each time step. The Chimera overlapped grid scheme allows subdomains move relative to each other, and provides a great flexibility when the boundary movement creates large displacements. The performance of the two time integration schemes for time-accurate computations is investigated. For an unsteady flow which requires small physical time step, the pressure projection method was found to be computationally efficient since it does not require any subiterations procedure. It was observed that the artificial compressibility method requires a fast convergence scheme at each physical time step in order to satisfy incompressibility condition. This was obtained by using a GMRES-ILU(0) solver in our computations. When a line-relaxation scheme was used, the time accuracy was degraded and time-accurate computations became very expensive. The current geometry for the LOX boost turbopump has various rotating and stationary components, such as inducer, stators, kicker, hydrolic turbine, where the flow is extremely unsteady. Figure 1 shows the geometry and computed surface pressure of the inducer. The inducer and the hydrolic turbine rotate in different rotational speed.

Kiris, Cetin↗

Enabling a Highly-Scalable Global Address Space Model for Petascale Computing

Over the past decade, the trajectory to the petascale has been built on increased complexity and scale of the underlying parallel architectures. Meanwhile, software de- velopers have struggled to provide tools that maintain the productivity of computational science teams using these new systems. In this regard, Global Address Space (GAS) programming models provide a straightforward and easy to use addressing model, which can lead to improved produc- tivity. However, the scalability of GAS depends directly on the design and implementation of the runtime system on the target petascale distributed-memory architecture. In this paper, we describe the design, implementation, and optimization of the Aggregate Remote Memory Copy Interface (ARMCI) runtime library on the Cray XT5 2.3 PetaFLOPs computer at Oak Ridge National Laboratory. We optimized our implementation with the flow intimation technique that we have introduced in this paper. Our optimized ARMCI implementation improves scalability of both the Global Arrays (GA) programming model and a real-world chemistry application NWChem from small jobs up through 180,000 cores.

97 MATHEMATICS AND COMPUTING↗