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 1,117 records · Page 62

Recovery Act: SeaMicro Volume Server Power Reduction Research Development

Cloud data centers are projected to be the fastest growing segment of the server market through 2015, according to IDC. Increasingly people and businesses rely on the Cloud to deliver digital content quickly and efficiently. Recovery Act funding from the Department of Energy has helped SeaMicro's technologies enhance the total cost of operation, performance and energy efficiency in large data center and Cloud environments. SeaMicro's innovative supercomputer fabric connects thousands of processor cores, memory, storage and input/output traffic. The company's fabric supports multiple processor instruction sets. Current systems featuring SeaMicro technology typically use one quarter the power and take one sixth the space of traditional servers with the same compute performance, yet deliver up to 12 times the bandwidth per core. Mozilla and eHarmony are two customers successfully using SeaMicro's technology. Numerous non-public customers have been successfully using the SeaMicro product in test and production facilities. As a result of the Recovery Act funding from the U.S. Department of Energy, more than 50 direct jobs were created at SeaMicro. To date, they primarily have been high-value, engineering jobs. Hardware, software and manufacturing engineering positions have been created, as well as sales and sales engineering. The positions have allowed SeaMicro to significantly accelerate engineering development and accelerate commercialization. As a result, commercialization and delivery to market are months ahead of initial schedule. Additional jobs were indirectly created through the development of the SeaMicro product. Through many years of research and hard work prior to receipt of public funding, SeaMicro was awarded 2 patents for its work. SeaMicro's product led the way for industry leaders to reconsider the market for low power servers and create new product lines. With valuable support of the U.S. Department of Energy and through SeaMicro's product, the market has been reshaped by the company's unique approach and ground breaking technology.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Electric field control of spin re-orientation in perpendicular magnetic tunnel junctions—CoFeB and MgO thickness dependence

We report an investigation of electric-field (EF) control of spin re-orientation as functions of the thicknesses of CoFeB free layer (FL) and MgO layer in synthetic-antiferromagnetic pinned magnetic tunnel junctions with perpendicular magnetic anisotropy. It is found that the EF modulates the coercivity (Hc) of the FL almost linearly for all FL thicknesses, while the EF efficiency, i.e., the slope of the linearity, increases as the FL thickness increases. This linear variation in Hc is also observed for larger MgO thicknesses (≥1.5 nm), while the EF efficiency increases only slightly from 370 to 410 Oe nm/V when MgO thickness increases from 1.5 to 1.76 nm. We have further observed the absence of quasi-DC unipolar switching. We discuss its origin and highlight the underlying challenges to implement the EF controlled switching in a practical magnetic memory.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Aerodynamic Shape Optimization of Supersonic Aircraft Configurations via an Adjoint Formulation on Parallel Computers

This work describes the application of a control theory-based aerodynamic shape optimization method to the problem of supersonic aircraft design. The design process is greatly accelerated through the use of both control theory and a parallel implementation on distributed memory computers. Control theory is employed to derive the adjoint differential equations whose solution allows for the evaluation of design gradient information at a fraction of the computational cost required by previous design methods. The resulting problem is then implemented on parallel distributed memory architectures using a domain decomposition approach, an optimized communication schedule, and the MPI (Message Passing Interface) Standard for portability and efficiency. The final result achieves very rapid aerodynamic design based on higher order computational fluid dynamics methods (CFD). In our earlier studies, the serial implementation of this design method was shown to be effective for the optimization of airfoils, wings, wing-bodies, and complex aircraft configurations using both the potential equation and the Euler equations. In our most recent paper, the Euler method was extended to treat complete aircraft configurations via a new multiblock implementation. Furthermore, during the same conference, we also presented preliminary results demonstrating that this basic methodology could be ported to distributed memory parallel computing architectures. In this paper, our concern will be to demonstrate that the combined power of these new technologies can be used routinely in an industrial design environment by applying it to the case study of the design of typical supersonic transport configurations. A particular difficulty of this test case is posed by the propulsion/airframe integration.

Reuther, James↗

Parallel DTFE Surface Density Field Reconstruction

We improve the interpolation accuracy and efficiency of the Delaunay tessellation field estimator (DTFE) for surface density field reconstruction by proposing an algorithm that takes advantage of the adaptive triangular mesh for line-of-sight integration. The costly computation of an intermediate 3D grid is completely avoided by our method and only optimally chosen interpolation points are computed, thus, the overall computational cost is significantly reduced. The algorithm is implemented as a parallel shared-memory kernel for large-scale grid rendered field reconstructions in our distributed-memory framework designed for N-body gravitational lensing simulations in large volumes. We also introduce a load balancing scheme to optimize the efficiency of processing a large number of field reconstructions. Our results show our kernel outperforms existing software packages for volume weighted density field reconstruction, achieving ~ 10x speedup, and our load balancing algorithm gains an additional ~ 3.6x speedup at scales with ~ 16k processes.

97 MATHEMATICS AND COMPUTING↗

An efficient approach to unstructured mesh hydrodynamics on the cell broadband engine

Unstructured mesh physics for the Cell Broadband Engine (CBE) has received little or no attention to date, largely because the CBE architecture poses particular challenges for unstructured mesh algorithms. The most common SPU memory management strategies cannot be applied to the irregular memory access patterns of unstructured meshes, and the SPU vector instruction set does not support the indirect addressing needed by connectivity arrays. This paper presents an approach to unstructured mesh physics that addresses these challenges, by creating a new mesh data structure and reorganizing code to give efficient CBE performance. The approach is demonstrated on the FLAG production hydrodynamics code using standard test problems, and results show an average speedup of more than 5x over the original code.

99 GENERAL AND MISCELLANEOUS↗

Efficient block preconditioned eigensolvers for linear response time-dependent density functional theory

We present two efficient iterative algorithms for solving the linear response eigen- value problem arising from the time dependent density functional theory. Although the matrix to be diagonalized is nonsymmetric, it has a special structure that can be exploited to save both memory and floating point operations. In particular, the nonsymmetric eigenvalue problem can be transformed into a product eigenvalue problem that is self-adjoint with respect to a K-inner product. This product eigenvalue problem can be solved efficiently by a modified Davidson algorithm and a modified locally optimal block preconditioned conjugate gradient (LOBPCG) algorithm that make use of the K-inner product. The solution of the product eigenvalue problem yields one component of the eigenvector associated with the original eigenvalue problem. However, the other component of the eigenvector can be easily recovered in a postprocessing procedure. Therefore, the algorithms we present here are more efficient than existing algorithms that try to approximate both components of the eigenvectors simultaneously. The efficiency of the new algorithms is demonstrated by numerical examples.

97 MATHEMATICS AND COMPUTING↗

GrainNN: A neighbor-aware long short-term memory network for predicting microstructure evolution during polycrystalline grain formation

High fidelity simulations of grain formation in alloys are an indispensable tool for process-to-mechanical-properties characterization. Such simulations, however, can be computationally expensive as they require fine spatial and temporal discretizations. Their cost becomes an obstacle to parametric studies and ensemble runs and ultimately makes downstream tasks like optimal control and uncertainty quantification challenging. To enable such downstream tasks, we introduce GrainNN, an efficient and accurate reduced-order model for epitaxial grain growth in additive manufacturing conditions. GrainNN is a sequence-to-sequence long-short-term-memory (LSTM) deep neural network that evolves the dynamics of manually crafted features. Its innovations are (1) an attention mechanism with grain-microstructure-specific transformer architecture; and (2) an overlapping combination of several clones of the network to generalize to grain configurations that are different from those used for training. This design enables GrainNN to predict grain formation for unseen physical parameters, grain number, domain size and geometry. Furthermore, GrainNN not only reconstructs the quantities of interest but also can be pointwise accurate. In our numerical experiments, we use a polycrystalline phase field method to both generate the training data and assess GrainNN. For multiparametric, ensemble simulations with many grains, GrainNN can be orders of magnitude faster than phase field simulations, while delivering 5%–15% pointwise error. Additionally, this speedup includes the cost of the phase field simulations for generating training data.

36 MATERIALS SCIENCE↗

Topological insulator-based high efficiency switching of magnetic unit, method and applications

A magneto-electronic device may include: a spin-orbit torque (SOT) generator layer; a magnetic memory layer; and/or sensing electrodes configured to measure a Hall effect of the magnetic memory layer. The SOT generator layer may include topological insulator material, and the magnetic memory layer may include ferromagnetic material with perpendicular magnetic anisotropy. A magneto-electronic device may include: a spin-orbit torque (SOT) generator layer; a first magnetic memory layer on the SOT generator layer; an insulating layer on the first magnetic memory layer; and/or a second magnetic memory layer on the insulating layer. The SOT generator layer may include topological insulator material. The first magnetic memory layer and the second magnetic memory layer may include ferromagnetic material with either perpendicular magnetic anisotropy or in-plane magnetic anisotropy.

36 MATERIALS SCIENCE↗

Cryogenic energy storage: Standalone design, rigorous optimization and techno-economic analysis

Energy storage allows flexible use and management of excess electricity and intermittently available renewable energy. Cryogenic energy storage (CES) is a promising storage alternative with a high technology readiness level and maturity, but the round-trip efficiency is often moderate and the Levelized Cost of Storage (LCOS) remains high. The complex flowsheets with intricate thermodynamics at cryogenic temperatures as well as the presence of multiple loops and refrigeration cycles pose considerable challenges for rigorous model-based design and optimization of CES systems. We present an optimization strategy that couples rigorous process simulation and Bayesian optimization with flowsheet decomposition and identification of hidden coupling constraints to optimally design standalone CES systems. Further refinement is done via a local search using the limited-memory Broyden–Fletcher–Goldfarb–Shanno algorithm. Here our results indicate that it is possible to achieve more than 52% round-trip efficiency and an LCOS of $153/MWh for a standalone 100 MW/400 MWh CES system limited to short-term storage with daily charging–discharging. However, a detailed techno-economic assessment reveals that the LCOS considering total capital investment may exceed $267/MWh when all direct and indirect costs of installation and operation are considered.

25 ENERGY STORAGE↗

Temporal compression of quantum-information-carrying photons using a photon-echo quantum memory approach

We study quantum compression and decompression of light pulses that carry quantum information using a photon-echo quantum memory technique with controllable inhomogeneous broadening of an isolated atomic absorption line. We investigate media with differently broadened absorption profiles, transverse and longitudinal, finding that the recall efficiency can be as large as unity and that the quantum information encoded into the photonic qubits can remain unperturbed. Our results provide insight into reversible light-atom interaction and are interesting in view of future quantum communication networks, where pulse compression and decompression may play an important role in increasing the qubit rate or in mapping quantum information from photonic carriers with large optical bandwidth into atomic memories with smaller bandwidth.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Aerodynamic Shape Optimization of Supersonic Aircraft Configurations via an Adjoint Formulation on Parallel Computers

This work describes the application of a control theory-based aerodynamic shape optimization method to the problem of supersonic aircraft design. The design process is greatly accelerated through the use of both control theory and a parallel implementation on distributed memory computers. Control theory is employed to derive the adjoint differential equations whose solution allows for the evaluation of design gradient information at a fraction of the computational cost required by previous design methods (13, 12, 44, 38). The resulting problem is then implemented on parallel distributed memory architectures using a domain decomposition approach, an optimized communication schedule, and the MPI (Message Passing Interface) Standard for portability and efficiency. The final result achieves very rapid aerodynamic design based on higher order computational fluid dynamics methods (CFD). In our earlier studies, the serial implementation of this design method (19, 20, 21, 23, 39, 25, 40, 41, 42, 43, 9) was shown to be effective for the optimization of airfoils, wings, wing-bodies, and complex aircraft configurations using both the potential equation and the Euler equations (39, 25). In our most recent paper, the Euler method was extended to treat complete aircraft configurations via a new multiblock implementation. Furthermore, during the same conference, we also presented preliminary results demonstrating that the basic methodology could be ported to distributed memory parallel computing architectures [241. In this paper, our concem will be to demonstrate that the combined power of these new technologies can be used routinely in an industrial design environment by applying it to the case study of the design of typical supersonic transport configurations. A particular difficulty of this test case is posed by the propulsion/airframe integration.

Reuther, James↗

NoC-enabled Software/Hardware Co-Design Framework for Accelerating k-mer Counting

Counting k-mers (substrings of fixed length k) in DNA and protein sequences generate non-uniform and irregular memory access patterns. Processing-in-Memory (PIM) architectures have potential to significantly reduce the overheads associated with such frequent and irregular memory accesses. However, existing k-mer counting algorithms are not designed to exploit the advantages of PIM architectures. Furthermore, owing to thermal constraints, the allowable power budget is limited in conventional PIM designs. Moreover, k-mer counting generates long-range, non-uniform traffic patterns which needs to be handled by an efficient Network-on-Chip (NoC). In this paper, we present an NoC-enabled hardware/software co-design framework to implement high-performance k-mer counting. The proposed architecture enables more computational power, efficient communication between cores/memory – all without creating a thermal bottleneck. The software component overcomes the I/O bottleneck, aids in NoC design and exposes more in-memory opportunities to exploit in PIM. Experimental results show that the proposed architecture outperforms a state-of-the-art software implementation using Hybrid Memory Cube (HMC), by up to 7.14X, while allowing significantly higher power budget.

Joardar, Biresh K.↗

Performance of hashed cache data migration schemes on multicomputers

After conducting an examination of several data-migration mechanisms which permit an explicit and controlled mapping of data to memory, a set of schemes for storage and retrieval of off-processor array elements is experimentally evaluated and modeled. All schemes considered have their basis in the use of hash tables for efficient access of nonlocal data. The techniques in question are those of hashed cache, partial enumeration, and full enumeration; in these, nonlocal data are stored in hash tables, so that the operative difference lies in the amount of memory used by each scheme and in the retrieval mechanism used for nonlocal data.

Hiranandani, Seema↗

Anodized Aluminum Oxide Membrane Ionic Memristors

Memory effect in ion transport (IT) at the solid–solution interface is uniquely attractive in that the conductance depends on or “memorizes” the previous states. Hysteretic and rectified transport properties offer exciting potential to developing advanced iontronics and neuromorphic functions, improving the efficiency of energy conversion and electrochemical processes, and overcoming the selectivity-throughput bottleneck in the enrichment of low abundant species for environment- and energy-friendly separations, among others. Herein, memory effects are discovered in the rectified electrokinetic IT through anodized aluminum oxide (AAO) membranes containing densely packed highly ordered nanochannels (10 10 per cm 2 ). Characteristic memristor responses of pinched current–potential loops are resolved in voltammetric experiments and successfully reproduced through finite element simulation. Excitatory and inhibitory conductance states are shown to arise from the enrichment and depletion of mobile charge carriers. Structurewise, the transport symmetry is broken by the barrier oxide layer (BOL) on the one end of the cylindrical nanochannels across the AAO membranes. Charge selectivity is attributed to the gradient(s) of the space charge density across the BOL characterized by depth profiling via X-ray photoelectron spectroscopy analysis. The space charge gradient(s) overcomes the fundamental limitation of widely exploited surface charge effects to enable intense rectification and hysteresis prevailing at very high ionic concentrations up to 1–2 M. A new strategy is developed for controlling the preferential IT direction and selectivity via counterion intercalation and extraction/exchange. Mechanistic understanding is further confirmed through parameter variations such as potential scan rate and ionic strength, which also demonstrates convenient controls of the related functions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

An autoencoder compression approach for accelerating large-scale inverse problems

Abstract Partial differential equation (PDE)-constrained inverse problems are some of the most challenging and computationally demanding problems in computational science today. Fine meshes required to accurately compute the PDE solution introduce an enormous number of parameters and require large-scale computing resources such as more processors and more memory to solve such systems in a reasonable time. For inverse problems constrained by time-dependent PDEs, the adjoint method often employed to compute gradients and higher order derivatives efficiently requires solving a time-reversed, so-called adjoint PDE that depends on the forward PDE solution at each timestep. This necessitates the storage of a high-dimensional forward solution vector at every timestep. Such a procedure quickly exhausts the available memory resources. Several approaches that trade additional computation for reduced memory footprint have been proposed to mitigate the memory bottleneck, including checkpointing and compression strategies. In this work, we propose a close-to-ideal scalable compression approach using autoencoders to eliminate the need for checkpointing and substantial memory storage, thereby reducing the time-to-solution and memory requirements. We compare our approach with checkpointing and an off-the-shelf compression approach on an earth-scale ill-posed seismic inverse problem. The results verify the expected close-to-ideal speedup for the gradient and Hessian-vector product using the proposed autoencoder compression approach. To highlight the usefulness of the proposed approach, we combine the autoencoder compression with the data-informed active subspace (DIAS) prior showing how the DIAS method can be affordably extended to large-scale problems without the need for checkpointing and large memory.

Mathematics↗

Computing rank‐revealing factorizations of matrices stored out‐of‐core

This paper describes efficient algorithms for computing rank-revealing factorizations of matrices that are too large to fit in main memory (RAM), and must instead be stored on slow external memory devices such as disks (out-of-core or out-of-memory). Traditional algorithms for computing rank-revealing factorizations (such as the column pivoted QR factorization and the singular value decomposition) are very communication intensive as they require many vector-vector and matrix-vector operations, which become prohibitively expensive when data is not in RAM. Randomization allows to reformulate new methods so that large contiguous blocks of the matrix are processed in bulk. The paper describes two distinct methods. The first is a blocked version of column pivoted Householder QR, organized as a “left-looking” method to minimize the number of the expensive write operations. The second method results employs a UTV factorization. It is organized as an algorithm-by-blocks to overlap computations and I/O operations. As it incorporates power iterations, it is much better at revealing the numerical rank. Numerical experiments on several computers demonstrate that the new algorithms are almost as fast when processing data stored on slow memory devices as traditional algorithms are for data stored in RAM.

97 MATHEMATICS AND COMPUTING↗

A splitting Chebyshev collocation method for Schrödinger–Poisson system

We develop a splitting Chebyshev collocation (SCC) method for the time-dependent Schrödinger–Poisson (SP) system arising from theoretical analysis of quantum plasmas. By means of splitting technique in time, the time-dependant SP system is first reduced to uncoupled Schrödinger and Poisson equations at every time step. The space variables in Schrödinger and Poisson equations are next represented by high-order Chebyshev polynomials, and the resulting system are discretized by the spectral collocation method. Finally, matrix diagonalization technique is applied to solve the fully discretized system in one dimension, two dimensions and three dimensions, respectively. The newly proposed method not only achieves spectral accuracy in space but also reduces the computer-memory requirements and the computational time in comparison with conventional solver. Numerical results confirm the spectral accuracy and efficiency of this method, and indicate that the SCC method could be an efficient alternative method for simulating the dynamics of quantum plasmas.

97 MATHEMATICS AND COMPUTING↗

Unconventional Quantum Advantages for Computation (U-QuAC)

While quantum computing offers the promise of exponential advantages, limited quantum speedups are known, especially for practical applications. To open new avenues for quantum advantages, we propose Unconventional Quantum Advantages for Computation (U-QuACs), with respect to unconventional resources such as space (number of bits or quantum bits of memory required to solve a problem), accuracy of solution, communication, or energy consumption. We focus on space-efficient quantum algorithms, where we seek to design algorithms that solve a problem using much less space than the total size of the input. A natural setting in which space is critical is the streaming model of computation, where the input data arrives sequentially in pieces that must each be processed individually. Streaming is motivated by a variety of problems including analysis of internet traffic or social networks. We design the first exponential quantum space advantage for a natural streaming problem, which also constitutes the first quantum advantage for approximating a discrete optimization problem, albeit with respect to space.

97 MATHEMATICS AND COMPUTING↗