Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallelization 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 1,567 records · Page 87

A 1D coupled transport/cold plasma wave model for parallel ponderomotive density modification near RF actuators

The plasma density profile close to RF actuators plays a vital role in the efficiency of power coupling from the antenna to the core plasma of fusion devices. Nonlinear interactions can be significant when considering the interaction of applied RF power and the tenuous SOL plasma, and the inclusion of this region in computational analysis is essential to providing robust predictive simulations of plasma response to RF application. Of particular interest is the observed density modification during RF operation. Here we present a self consistent description of the nonlinear ponderomotive force in the antenna near field region, via coupling a 1D field aligned transport model with a 1D cold plasma wave model. Results are presented, which show a parallel electric field of 300 Vm -1 drove a density reduction of ~ 20% over 25 RF periods.

Barnett, Rhea↗

Partitioning problems in parallel, pipelined, and distributed computing

The problem of optimally assigning the modules of a parallel program over the processors of a multiple-computer system is addressed. A sum-bottleneck path algorithm is developed that permits the efficient solution of many variants of this problem under some constraints on the structure of the partitions. In particular, the following problems are solved optimally for a single-host, multiple-satellite system: partitioning multiple chain-structured parallel programs, multiple arbitrarily structured serial programs, and single-tree structured parallel programs. In addition, the problem of partitioning chain-structured parallel programs across chain-connected systems is solved under certain constraints. All solutions for parallel programs are equally applicable to pipelined programs. These results extend prior research in this area by explicitly taking concurrency into account and permit the efficient utilization of multiple-computer architectures for a wide range of problems of practical interest.

Bokhari, Shahid H.↗

Real-time optimization of multi-cell industrial evaporative cooling towers using machine learning and particle swarm optimization

Existing electrical generating stations must operate with greater flexibility due to increasing renewable energy penetration on the electrical grid, and many coal-fired power stations have transitioned away from baseload operation to load-following operation to aid in grid stability. In cases where multiple independently controlled cooling tower cells are used in parallel for the cooling purposes of such stations, there is an opportunity to increase plant efficiency through data-driven optimization across their full load ranges. This work presents a novel application of real-time optimization using machine learning and particle swarm optimization on a multi-cell induced-draft cooling tower servicing a coal-fired power station under variable load. This is the first work to demonstrate simultaneous optimization of a multi-cell cooling tower, in addition to using machine learning for closed-loop control on a cooling tower. A novel control configuration is presented that ensures original control logic is not adversely affected and that the overall plant process is not disrupted using only existing hardware and operational data. To verify this methodology, the 12 independent cooling tower cells are simulated in parallel using historic operating data to demonstrate the effectiveness of real-time optimization compared to current practice. An artificial neural network is trained to predict overall cooling tower power consumption using only operational data and ambient conditions with an R2 value of greater than 0.96. The real-time optimization using particle swarm yields 6.7% annual energy usage savings compared to current practices, although the extent of the real-time savings varies greatly with both plant load and environmental conditions. This is particularly significant for a variable load situation because frequent ramping typically results in reduced overall efficiency. Furthermore, this proposed AI-based solution presents an opportunity to improve the overall heat rate of a load-following coal-fired power plant without the need to perform extensive first-principles modeling or add additional hardware to the cooling tower, resulting in more resources conserved and less overall emissions per unit of electricity generated.

42 ENGINEERING↗

Mathematical and Numerical Aspects of the Adaptive Fast Multipole Poisson-Boltzmann Solver

This paper summarizes the mathematical and numerical theories and computational elements of the adaptive fast multipole Poisson-Boltzmann (AFMPB) solver. We introduce and discuss the following components in order: the Poisson-Boltzmann model, boundary integral equation reformulation, surface mesh generation, the nodepatch discretization approach, Krylov iterative methods, the new version of fast multipole methods (FMMs), and a dynamic prioritization technique for scheduling parallel operations. For each component, we also remark on feasible approaches for further improvements in efficiency, accuracy and applicability of the AFMPB solver to large-scale long-time molecular dynamics simulations. Lastly, the potential of the solver is demonstrated with preliminary numerical results.

97 MATHEMATICS AND COMPUTING↗

Improvement of force factor of magnetostrictive vibration power generator for high efficiency

We develop high power magnetostrictive vibration power generator for battery-free wireless electronics. The generator is based on a cantilever of parallel beam structure consisting of coil-wound Galfenol and stainless plates with permanent magnet for bias. Oscillating force exerted on the tip bends the cantilever in vibration yields stress variation of Galfenol plate, which causes flux variation and generates voltage on coil due to the law of induction. This generator has advantages over conventional, such as piezoelectric or moving magnet types, in the point of high efficiency, highly robust, and low electrical impedance. Our concern is the improvement of energy conversion efficiency dependent on the dimension. Especially, force factor, the conversion ratio of the electromotive force (voltage) on the tip velocity in vibration, has an important role in energy conversion process. First, the theoretical value of the force factor is formulated and then the validity was verified by experiments, where we compare four types of prototype with parameters of the dimension using 7.0 × 1.5 × 50 mm beams of Galfenol with 1606-turn wound coil. In addition, the energy conversion efficiency of the prototypes depending on load resistance was measured. The most efficient prototype exhibits the maximum instantaneous power of 0.73 W and energy of 4.7 mJ at a free vibration of frequency of 202 Hz in the case of applied force is 25 N. Further, it was found that energy conversion efficiency depends not only on the force factor but also on the damping (mechanical loss) of the vibration.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Rigorous justification for the space–split sensitivity algorithm to compute linear response in Anosov systems

Ruelle gave a formula for linear response of transitive Anosov diffeomorphisms. Recently, practically computable realizations of Ruelle's formula have emerged that potentially enable sensitivity analysis of certain high-dimensional chaotic numerical simulations encountered in the applied sciences. In this paper, we provide full mathematical justification for the convergence of one such efficient computation, the space–split sensitivity, or S3, algorithm. In S3, Ruelle's formula is computed as a sum of two terms obtained by decomposing the perturbation vector field into a coboundary and a remainder that is parallel to the unstable direction. Such a decomposition results in a splitting of Ruelle's formula that is amenable to efficient computation. Here, we prove the existence of the S3 decomposition and the convergence of the computations of both resulting components of Ruelle's formula.

97 MATHEMATICS AND COMPUTING↗

Efficient discontinuous finite difference meshes for 3-D Laplace–Fourier domain seismic wavefield modelling in acoustic media with embedded boundaries

Simulation of acoustic wave propagation in the Laplace–Fourier (LF) domain, with a spatially uniform mesh, can be computationally demanding especially in areas with large velocity contrasts. To improve efficiency and convergence, we use 3-D second- and fourth-order velocity-pressure finite difference (FD) discontinuous meshes (DM). Our DM algorithm can use any spatial discretization ratio between meshes. We evaluate direct and iterative parallel solvers for computational speed, memory requirements and convergence. Finally, benchmarks in realistic 3-D models and topographies show more efficient and stable results for DM with direct solvers than uniform mesh results with iterative solvers.

58 GEOSCIENCES↗

GPU Implementation of the OVERFLOW CFD Code

The high-performance computing (HPC) landscape is quickly changing to systems where most of the performance comes from specialized chips, specifically graphics processing units (GPUs). Such GPU systems are throughput machines, where efficient use of the GPU often requires code refactoring to expose a few orders of magnitude more fine grain parallelism than was previously used on the CPU. Recent modifications to OVERFLOW, an overset, structured grid, computational fluid dynamics flow solver, written in Fortran will be presented. These modifications include both code modernization efforts and algorithmic changes to enable OVERFLOW to efficiently utilize GPUs. Many of these algorithmic changes would likely also be applicable for other structured grid, stencil-based codes wanting to utilize GPUs. The capabilities that have been ported to run on the GPUs are presented, along with the performance gains of the GPU version relative the CPU version of OVERFLOW.

GPU Programming↗

GPU Implementation of the OVERFLOW CFD Code

The high-performance computing (HPC) landscape is quickly changing to systems where most of the performance comes from specialized chips, specifically graphics processing units (GPUs). Such GPU systems are throughput machines, where efficient use of the GPU often requires code refactoring to expose a few orders of magnitude more fine grain parallelism than was previously used on the CPU. Recent modifications to OVERFLOW, an overset, structured grid, computational fluid dynamics flow solver, written in Fortran will be presented. These modifications include both code modernization efforts and algorithmic changes to enable OVERFLOW to efficiently utilize GPUs. Many of these algorithmic changes would likely also be applicable for other structured grid, stencil-based codes wanting to utilize GPUs. The capabilities that have been ported to run on the GPUs are presented, along with the performance gains of the GPU version relative the CPU version of OVERFLOW.

GPU Programming↗

Algorithms and analyses for stochastic optimization for turbofan noise reduction using parallel reduced-order modeling

Simulation-based optimization of acoustic liner design in a turbofan engine nacelle for noise reduction purposes can dramatically reduce the cost and time needed for experimental designs. Because uncertainties are inevitable in the design process, a stochastic optimization algorithm is posed based on the conditional value-at-risk measure so that an ideal acoustic liner impedance is determined that is robust in the presence of uncertainties. A parallel reduced-order modeling framework is developed that dramatically improves the computational efficiency of the stochastic optimization solver for a realistic nacelle geometry. The reduced stochastic optimization solver takes less than 500 s to execute. Additionally, well-posedness and finite element error analyses of the state system and optimization problem are provided.

97 MATHEMATICS AND COMPUTING↗

Optimizing checkpoint data placement with guaranteed burst buffer endurance in large-scale hierarchical storage systems

Non-volatile devices, such as SSDs, will be an integral part of the deepening storage hierarchy on large-scale HPC systems. These devices can be on the compute nodes as part of a distributed burst buffer service or they can be external. Wherever they are located in the hierarchy, one critical design issue is the SSD endurance under the write-heavy workloads, such as the checkpoint I/O for scientific applications. For these environments, it is widely assumed that checkpoint operations can occur once every 60 min and for each checkpoint step as much as half of the system memory can be written out. Unfortunately, for large-scale HPC applications, the burst buffer SSDs can be worn out much more quickly given the extensive amount of data written at every checkpoint step. One possible solution is to control the amount of data written by reducing the checkpoint frequency. However, a direct effect caused by reduced checkpoint frequency is the increased vulnerability window of system failures and therefore potentially wasted computation time, especially for large-scale compute jobs.

Computer Science↗

Tailorable Cellulose II Nanocrystals (CNC II) Prepared in Mildly Acidic Lithium Bromide Trihydrate (MALBTH)

Preparing cellulose II nanocrystals (CNC II) requires a polymorph transformation of natural cellulose I feedstocks. The transformation is usually achieved via a process such as mercerization or dissolution–regeneration. This study demonstrated a new method to prepare CNC II directly from bleached kraft pulp (BKP, a commercially available cellulose I feedstock) in a mildly acidic lithium bromide trihydrate (MALBTH) system, a concentrated (~61 wt%) solution of LiBr in water with a very low concentration (2.5 mM) of sulfuric acid. First, the BKP was treated in the MALBTH system to generate a cellulose II hydrolysis solid residue (CHR) with a yield of 64–86%, during which the selective hydrolysis of disordered cellulose and the polymorph transformation were completed simultaneously. Then, subsequent oxidation of the CHR by ammonium persulfate (APS, 0.1–0.6 M) resulted in the CNC II with high yield (up to 62%), high crystallinity (over 90%), rich surface carboxyl group (0.3–1.2 mmol g-1 cellulose), excellent colloidal stability (up to -59 mV zeta potential), and high thermal stability. The CNC II had a tunable length (26–57 nm), determined by the conditions of the MALBTH hydrolysis and the APS oxidation, but similar lateral dimension (8–10 nm). The characterization of the CHR by wide-angle X-ray diffraction and Fourier transform infrared spectroscopy verified the polymorphic transformation from cellulose I to II during the MALBTH treatment. The swelling of the BKP in the MALBTH enabled cellulose crystallites to slide and reassemble, which completed the rearrangement of cellulose chains from parallel to anti-parallel conformation (polymorph transformation from cellulose I to II). This study provided an efficient and green method to produce cellulose II nanocrystals with controllable aspect ratios via the simultaneous hydrolysis and polymorph transformation of cellulose I feedstocks.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Method for cooling nanostructures to microkelvin temperatures

We propose a new scheme aimed at cooling nanostructures to microkelvin temperature based on the well established technique of adiabatic nuclear demagnetization: we attach each device measurement lead to an individual nuclear refrigerator, allowing efficient thermal contact to a microkelvin bath. On a prototype consisting of a parallel network of nuclear refrigerators, temperatures of {approx}1 mK simultaneously on ten measurement leads have been reached upon demagnetization, thus completing the first steps toward ultracold nanostructures.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Magnetization switching of a metallic nanomagnet via current-induced surface spin-polarization of an underlying topological insulator

We consider a thermally stable, metallic nanoscale ferromagnet (FM) subject to spin-polarized current injection and exchange coupling from the spin-helically locked surface states of a topological insulator (TI) to evaluate possible non-volatile memory applications. We consider parallel transport in the TI and the metallic FM, and focus on the efficiency of magnetization switching as a function of transport between the TI and the FM. Transport is modeled as diffusive in the TI beneath the FM, consistent with the mobility in the TI at room temperature, and in the FM, which essentially serves as a constant potential region albeit spin-dependent except in the low conductivity, diffusive limit. Thus, it can be captured by drift-diffusion simulation, which allows for ready interpretation of the results. We calculate switching time and energy consumed per write operation using self-consistent transport, spin-transfer-torque (STT), and magnetization dynamics calculations. Calculated switching energies and times compare favorably to conventional spin-torque memory schemes for substantial interlayer conductivity. Nevertheless, we find that shunting of current from the TI to a metallic nanomagnet can substantially limit efficiency. Exacerbating the problem, STT from the TI effectively increases the TI resistivity. We show that for optimum performance, the sheet resistivity of the FM layer should be comparable to or larger than that of the TI surface layer. Thus, the effective conductivity of the FM layer becomes a critical design consideration for TI-based non-volatile memory.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Enhancing high-fidelity neural network potentials through low-fidelity sampling

The efficacy of neural network potentials (NNPs) critically depends on the quality of the configurational datasets used for training. Prior research using empirical potentials has shown that well-selected liquid–solid transitional configurations of a metallic system can be translated to other metallic systems. This study demonstrates that such validated configurations can be relabeled using density functional theory (DFT) calculations, thereby enhancing the development of high-fidelity NNPs. Training strategies and sampling approaches are efficiently assessed using empirical potentials and subsequently relabeled via DFT in a highly parallelized fashion for high-fidelity NNP training. Our results reveal that relying solely on energy and force for NNP training is inadequate to prevent overfitting, highlighting the necessity of incorporating stress terms into the loss functions. To optimize training involving force and stress terms, we propose employing transfer learning to fine-tune the weights, ensuring that the potential surface is smooth for these quantities composed of energy derivatives. This approach markedly improves the accuracy of elastic constants derived from simulations in both empirical potential-based NNPs and relabeled DFT-based NNPs. Overall, this study offers significant insights into leveraging empirical potentials to expedite the development of reliable and robust NNPs at the DFT level.

97 MATHEMATICS AND COMPUTING↗

THE ROLE OF PRESSURE ANISOTROPY ON PARTICLE ACCELERATION DURING MAGNETIC RECONNECTION

Voyager spacecraft observations have revealed that contrary to expectations, the source of anomalous cosmic rays (ACRs) is not at the local termination shock. A possible mechanism of ACR acceleration is magnetic reconnection in the heliosheath. Using a particle-in-cell code, we investigate the effects of {beta} on reconnection-driven particle acceleration by studying island growth in multiple interacting Harris current sheets. Many islands are generated, and particles are dominantly heated through Fermi reflection in contracting islands during island growth and merging. There is a striking difference between the heating of electrons versus the heating of ions. There is a strong dependence of {beta} on electron heating, while the ion heating is insensitive to {beta}. Anisotropies develop with T {sub Parallel-To} {ne} T for both electrons and ions. The electron anisotropies support the development of a Weibel instability that suppresses the Fermi acceleration of the electrons. Since the Weibel instability develops at a larger T {sub Parallel-To }/T in lower {beta} systems, electrons are able to accelerate more efficiently by the Fermi mechanism at low {beta}. The variance in anisotropy implies less electron acceleration in higher {beta} systems, and thus less heating. This study sheds light on particle acceleration mechanisms within the sectored magnetic field regions of the heliosheath and the dissipation of turbulence such as that produced by the magnetorotational instability in accreting systems.

79 ASTRONOMY AND ASTROPHYSICS↗

Enhancing lipid production in plant cells through automated high-throughput genome engineering and phenotyping

Plant bioengineering is a time-consuming and labor-intensive process with no guarantee of achieving desired traits. Here, we present a fast, automated, scalable, high-throughput pipeline for plant bioengineering (FAST-PB) in maize (Zea mays) and Nicotiana benthamiana. FAST-PB enables genome editing and product characterization by integrating automated biofoundry engineering of callus and protoplast cells with single-cell matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). We first demonstrated that FAST-PB could streamline Golden Gate cloning, with the capacity to construct 96 vectors in parallel. Using FAST-PB in protoplasts, we found that PEG2050 increased transfection efficiency by over 45%. For proof-of-concept, we established a reporter-gene-free method for CRISPR editing and phenotyping via mutation of high chlorophyll fluorescence 136. We show that diverse lipids were enhanced up to 6-fold using CRISPR activation of lipid controlling genes. In callus cells, an automated transformation platform was employed to regenerate plants with enhanced lipid traits through introducing multigene cassettes. Lastly, FAST-PB enabled high-throughput single-cell lipid profiling by integrating MALDI-MS with the biofoundry, protoplast, and callus cells, differentiating engineered and unengineered cells using single-cell lipidomics. Furthermore, these innovations massively increase the throughput of synthetic biology, genome editing, and metabolic engineering and change what is possible using single-cell metabolomics in plants.

59 BASIC BIOLOGICAL SCIENCES↗

Anomalous thermal effect in Zr Te 5 observed via photothermal measurements

In this study, we explore the magneto-thermoelectric power (MTP) of $ZrTe$ 5 , a canonical Dirac semimetal, through a novel photothermal technique. Unlike conventional thermoelectric studies that rely on on-chip heaters and are limited by fabrication processes, especially for stress-sensitive materials, our approach utilizes photothermal effects to induce temperature gradients. Our experiments, applying a magnetic field approximately parallel and transverse to the photocurrent detection direction, reveal that the photothermal method efficiently and reliably extracts both diagonal and off-diagonal components of the thermoelectric coefficient of $ZrTe$ 5 . Here, we observe that the longitudinal MTP reproduces features previously reported in thermal transport studies, while the photoinduced transverse MTP confirms the anomalous Nernst effect. This photothermal measurement technique opens new avenues for investigating transport properties in a wide range of quantum materials, both in 3D and 2D systems.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗