Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallel processing”

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 955 records · Page 53

P3DFFT: A Framework for Parallel Computations of Fourier Transforms in Three Dimensions

Fourier and related transforms are a family of algorithms widely employed in diverse areas of computational science, notoriously difficult to scale on high-performance parallel computers with a large number of processing elements (cores). This paper introduces a popular software package called P3DFFT which implements fast Fourier transforms (FFTs) in three dimensions in a highly efficient and scalable way. It overcomes a well-known scalability bottleneck of three-dimensional (3D) FFT implementations by using two-dimensional domain decomposition. Designed for portable performance, P3DFFT achieves excellent timings for a number of systems and problem sizes. On a Cray XT5 system P3DFFT attains 45% efficiency in weak scaling from 128 to 65,536 computational cores. Library features include Fourier and Chebyshev transforms, Fortran and C interfaces, in- and out-of-place transforms, uneven data grids, and single and double precision. P3DFFT is available as open source at http://code.google.com/p/p3dfft/. This paper discusses P3DFFT implementation and performance in a way that helps guide the user in making optimal choices for parameters of their runs. Read More: https://epubs.siam.org/doi/10.1137/11082748X

Mathematics↗

MPI on millions of cores

Petascale parallel computers with more than a million processing cores are expected to be available in a couple of years. Although MPI is the dominant programming interface today for large-scale systems that at the highest end already have close to 300,000 processors, a challenging question to both researchers and users is whether MPI will scale to processor and core counts in the millions. In this paper, we examine the issue of scalability of MPI to very large systems. We first examine the MPI specification itself and discuss areas with scalability concerns and how they can be overcome. We then investigate issues that an MPI implementation must address in order to be scalable. To illustrate the issues, we ran a number of simple experiments to measure MPI memory consumption at scale up to 131,072 processes, or 80%, of the IBM Blue Gene/P system at Argonne National Laboratory. Based on the results, we identified nonscalable aspects of the MPI implementation and found ways to tune it to reduce its memory footprint. We also briefly discuss issues in application scalability to large process counts and features of MPI that enable the use of other techniques to alleviate scalability limitations in applications.

97 MATHEMATICS AND COMPUTING↗

Hydration Mechanisms in Nanoparticle Interaction and Surface Energetics

Work on the project advanced molecular level understanding, prediction and control of nanoscale hydration in salt solutions under electric stimuli or ionic patterning. Permeation of nanoporous electrodes at preset voltage underlies the function of ultracapacitors. Transitory regulation of wetting in nanoporous media by electric field spans an array of applications in materials, energy storage, and separation sciences. The physically related modulation of nanoparticle solubility by surface charges can significantly extend the range of the nanomaterial applications, improve processing techniques, and potentially alleviate environmental concerns. Pore/solution equilibria, and activated kinetics of liquid gating by nanoconfined electrolytes are challenging problems at the forefront of experimental and theoretical research. Addressing these problems from a molecular perspective required the development of state-of-the-art simulation algorithms in statistical mechanics to capture complex processes in open systems under electric control. Parallel studies of wetting and dispersibility of polar and ionizing particles aim to uncover predictive relations between electrowetting, chemical functionalization, and geometry of nanomaterial particles. By nonequilibrium dynamic modeling, we elucidated electrolyte flow in nanochanels and associated electrokinetic energy conversion. Research on the program provided training opportunities for the next generation of scientists in computational chemistry. Insights, and methods from the project implicate broad segment of researchers in energy and nanosciences, materials and surface chemistry.

74 ATOMIC AND MOLECULAR PHYSICS↗

Airborne LiDAR to Improve Canopy Fuels Mapping for Wildfire Modeling

Increasing conflict between wildfire and the built environment has increased the need for more up-to-date and finer resolution canopy fuels data to improve wildfire modeling and associated risk forecasts. The US Forest Service and US Department of the Interior’s LANDFIRE product, which provides 30-m resolution canopy fuels data for the entire US, is one of the most widely used sources of fuels data. However, the last complete mapping effort for LANDFIRE is based on 2016 conditions, and subsequent updates reflect disturbances 1-2 years behind the release year. Airborne systems equipped with Light Detection and Ranging (LiDAR) sensors can be deployed to actively sense canopy structure and estimate canopy fuels data (cover, height, base height, bulk density) at finer resolutions. Canopy base height (CBH) and canopy bulk density (CBD) are difficult to measure both in the field and in LiDAR point clouds. Still, they are important for accurately modeling crown fires, which are often intense and difficult to contain. Additionally, point cloud datasets are large, and calculations require efficient utilization of computational resources. To address these challenges, we are working on an approach that uses openly available National Ecological Observatory Network (NEON) airborne LiDAR data, with calculations processed in the R programming language and parallelized through the lidR package. CBH and CBD are often derived from tree height, diameter at breast height, and species-specific allometries using the Fire and Fuels Extension of the Forest Vegetation Simulator (FFE-FVS). We aim to test if airborne LiDAR can estimate CBH and CBD without the use of empirical equations. Reliable estimates of canopy fuels data directly from airborne LiDAR could streamline quick, fine-resolution updates for use in wildfire behavior models.

54 ENVIRONMENTAL SCIENCES↗

Predicting Selective Laser Printing Print Quality of Polymer Powders through Melt Flow Index

Selective Laser Sintering (SLS) uses a precisely controlled laser to fuse polymer powder to build complex 3D shapes. While SLS covers a wide application space, the processing knowledge of polymer powder is limited, restricting the number of commercial powders available. This study serves to expand the processing knowledge of polypropylene and polyethylene in parallel with a “mature” SLS feedstock, nylon, through melt flow index (MFI) characterization. Differential Scanning Calorimetry (DSC) was used to explore the sintering window of the polymers. Polyethylene and polypropylene exhibited a relatively narrow sintering window between 4 – 5 °C, whereas the sintering window of nylon was much wider, between 23 – 24 °C. This suggests that the print quality between polyethylene and polypropylene would be similar; however, X-ray computed tomography revealed a higher volume of print defects, voids, and de lamination in polyethylene than in polypropylene. MFI analysis provided additional insight into the difference in print quality, as the MFI of polyethylene was 7.16 g/10 min, 9.95 g/10 min for polypropylene, and 17.23 g/10 min for nylon. MFI is inversely correlated to melt viscosity, and a low melt viscosity is desired for proper coalescing between the layers and particles. These results suggest that MFI is a promising tool, in conjunction with traditional thermal analysis, for screening candidate powder feedstocks for SLS and optimization of print parameters for novel powders.

36 MATERIALS SCIENCE↗

Accelerating the Computation of Detailed Chemical Reaction Kinetics for Simulating Combustion of Complex Fuels

Combustion of hydrocarbon fuels has been a very challenging scientific and engineering problem due to the complexity of turbulent flows and hydrocarbon reaction kinetics. There is an urgent need to develop an efficient modeling capability to accurately predict the combustion of complex fuels. Detailed chemical kinetic models for the surrogates of fuels such as gasoline, diesel and JP-8 consist of thousands of chemical species and Arrhenius reaction steps. Oxygenated fuels such as bio-fuels and heavier hydrocarbons, such as from newer fossil fuel sources, are expected to have a much more complex chemistry requiring increasingly larger chemical kinetic models. Such models are beyond current computational capability, except for homogeneous or partially stirred reactor type calculations. The advent of highly parallel multi-core processors and graphical processing units (GPUs) promises a steep increase in computational performance in the coming years. This paper will present a software framework that translates the detailed chemical kinetic models to high-performance code targeted for GPU accelerators.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Hybrid Parallelization of Adaptive MHD-Kinetic Module in Multi-Scale Fluid-Kinetic Simulation Suite

The Multi-Scale Fluid-Kinetic Simulation Suite has a computational tool set for solving partially ionized flows. In this paper we focus on recent developments of the kinetic module which solves the Boltzmann equation using the Monte-Carlo method. The module has been recently redesigned to utilize intra-node hybrid parallelization. We describe in detail the redesign process, implementation issues, and modifications made to the code. Finally, we conduct a performance analysis.

97 MATHEMATICS AND COMPUTING↗

The hydrogen tunneling splitting in malonaldehyde: A full-dimensional time-independent quantum mechanical method

The accurate time-independent quantum dynamics calculations on the ground-state tunneling splitting of malonaldehyde in full dimensionality are reported for the first time. This is achieved with an efficient method developed by us. In our method, the basis functions are customized for the hydrogen transfer process which has the effect of greatly reducing the size of the final Hamiltonian matrix, and the Lanczos method and parallel strategy are used to further overcome the memory and central processing unit time bottlenecks. The obtained ground-state tunneling splitting of 24.5 cm{sup −1} is in excellent agreement with the benchmark value of 23.8 cm{sup −1} computed with the full-dimensional, multi-configurational time-dependent Hartree approach on the same potential energy surface, and we estimate that our reported value has an uncertainty of less than 0.5 cm{sup −1}. Moreover, the role of various vibrational modes strongly coupled to the hydrogen transfer process is revealed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Design Considerations for GPU-based Mixed Integer Programming on Parallel Computing Platforms

Mixed Integer Programming (MIP) is a powerful abstraction in combinatorial optimization that finds real-life application across many significant sectors. The recent proliferation of graphical processing unit (GPU)-based accelerated computing architectures in large-scale parallel computing or supercomputing presents new opportunities as well as challenges in the advancement of MIP solver technology to effectively use the new accelerated computing platforms and scale to large parallel systems. Here, we recount the conventional processor-based strategies and focus on configurations where the most promising intersection lies between parallel MIP solver approaches and the specific strengths of accelerated parallel platforms. We note that the best potential lies in solving problems whose individual matrix sizes (of the linear program relaxation) fit entirely within one accelerator's memory and whose branch-and-bound (or branch-and-cut) trees cannot be fully contained within a small number of computational nodes. Additionally, we identify ideal features of computational linear algebra support on GPU accelerators that would help advance this direction of scalable parallel solution of MIP problems on GPU-based accelerated computing architectures.

Perumalla, Kalyan↗

Counting Kmers for Biological Sequences at Large Scale

Counting the abundance of all the distinct kmers in biological sequence data is a fundamental step in bioinformatics. These applications include de novo genome assembly, error correction, etc. With the development of sequencing technology, the sequence data in a single project can reach Petabyte-scale or Terabyte-scale nucleotides. Counting demand for the abundance of these sequencing data is beyond the memory and computing capacity of single computing node, and how to process it efficiently is a challenge on a high-performance computing cluster. As such, we propose SWAPCounter, a highly scalable distributed approach for kmer counting. This approach is embedded with an MPI streaming I/O module for loading huge data set at high speed, and a counting bloom filter module for both memory and communication efficiency. By overlapping all the counting steps, SWAPCounter achieves high scalability with high parallel efficiency. In this work, the experimental results indicate that SWAPCounter has competitive performance with two other tools on shared memory environment, KMC 2 , and MSPKmerCounter. Moreover, SWAPCounter also shows the highest scalability under strong scaling experiments. In our experiment on Cetus supercomputer, SWAPCounter scales to 32,768 cores with 79% parallel efficiency (using 2048 cores as baseline) when processing 4 TB sequence data of 1000 Genomes. The source code of SWAPCounter is publicly available at https://github.com/mengjintao/SWAPCounter.

59 BASIC BIOLOGICAL SCIENCES↗

Fast optimization and dose calculation in scanned ion beam therapy

Purpose: Particle therapy (PT) has advantages over photon irradiation on static tumors. An increased biological effectiveness and active target conformal dose shaping are strong arguments for PT. However, the sensitivity to changes of internal geometry complicates the use of PT for moving organs. In case of interfractionally moving objects adaptive radiotherapy (ART) concepts known from intensity modulated radiotherapy (IMRT) can be adopted for PT treatments. One ART strategy is to optimize a new treatment plan based on daily image data directly before a radiation fraction is delivered [treatment replanning (TRP)]. Optimizing treatment plans for PT using a scanned beam is a time consuming problem especially for particles other than protons where the biological effective dose has to be calculated. For the purpose of TRP, fast optimization and fast dose calculation have been implemented into the GSI in-house treatment planning system (TPS) TRiP98. Methods: This work reports about the outcome of a code analysis that resulted in optimization of the calculation processes as well as implementation of routines supporting parallel execution of the code. To benchmark the new features, the calculation time for therapy treatment planning has been studied. Results: Compared to the original version of the TPS, calculation times for treatment planning (optimization and dose calculation) have been improved by a factor of 10 with code optimization. The parallelization of the TPS resulted in a speedup factor of 12 and 5.5 for the original version and the code optimized version, respectively. Hence the total speedup of the new implementation of the authors' TPS yielded speedup factors up to 55. Conclusions: The improved TPS is capable of completing treatment planning for ion beam therapy of a prostate irradiation considering organs at risk in this has been overseen in the review process. Also see below 6 min.

60 APPLIED LIFE SCIENCES↗

Lignin-Derived Chemicals from Pulping Liquors: Cooperative Research and Development CRADA Number CRD-17-00714 (Final Report)

National Renewable Energy Laboratory (NREL) has developed a robust portfolio for lignin valorization using a “biological funneling” approach, which adopts a microbial process to convert the inherent heterogeneity of lignin-rich streams into single high-value products. In parallel, Sustainable Fiber Technologies (SFT) has developed a proprietary, high-pH process to convert non-woody biomass into valuable pulp for myriad biomaterial applications. Therefore, by combining the expertise of both SFT and NREL, this project aimed to valorize co-products from an industrial pulping process of non-woody feedstocks. A detailed analytical characterization of an SFT pulping stream was used to aid the development of a bench-scale separations strategy to isolate a mixture of low molecular weight compounds, and ultimately to aid in bioreactor cultivations used to produce polyhydroxyalkanoate (PHA) biopolymers at titers, rates, and yields necessary for moving towards commercialization.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Towards Scalable Deep Learning via I/O Analysis and Optimization

Deep learning systems have been growing in prominence as a way to automatically characterize objects, trends, and anomalies. Given the importance of deep learning systems, researchers have been investigating techniques to optimize such systems. An area of particular interest has been using large supercomputing systems to quickly generate effective deep learning networks: a phase often referred to as “training” of the deep learning neural network. As we scale existing deep learning frameworks—such as Caffe—on these large supercomputing systems, we notice that the parallelism can help improve the computation tremendously, leaving data I/O as the major bottleneck limiting the overall system scalability. In this paper, we first present a detailed analysis of the performance bottlenecks of Caffe on large supercomputing systems. Our analysis shows that the I/O subsystem of Caffe—LMDB—relies on memory-mapped I/O to access its database, which can be highly inefficient on large-scale systems because of its interaction with the process scheduling system and the network-based parallel filesystem. Based on this analysis, we then present LMDBIO, our optimized I/O plugin for Caffe that takes into account the data access pattern of Caffe in order to vastly improve I/O performance. Our experimental results show that LMDBIO can improve the overall execution time of Caffe by nearly 20-fold in some cases.

Pumma, Sarunya↗

THE MECHANISMS OF ELECTRON ACCELERATION DURING MULTIPLE X LINE MAGNETIC RECONNECTION WITH A GUIDE FIELD

The interactions between magnetic islands are considered to play an important role in electron acceleration during magnetic reconnection. In this paper, two-dimensional particle-in-cell simulations are performed to study electron acceleration during multiple X line reconnection with a guide field. Because the electrons remain almost magnetized, we can analyze the contributions of the parallel electric field, Fermi, and betatron mechanisms to electron acceleration during the evolution of magnetic reconnection through comparison with a guide-center theory. The results show that with the magnetic reconnection proceeding, two magnetic islands are formed in the simulation domain. Next, the electrons are accelerated by both the parallel electric field in the vicinity of the X lines and the Fermi mechanism due to the contraction of the two magnetic islands. Then, the two magnetic islands begin to merge into one, and, in such a process, the electrons can be accelerated by both the parallel electric field and betatron mechanisms. During the betatron acceleration, the electrons are locally accelerated in the regions where the magnetic field is piled up by the high-speed flow from the X line. At last, when the coalescence of the two islands into one big island finishes, the electrons can be further accelerated by the Fermi mechanism because of the contraction of the big island. With the increase of the guide field, the contributions of the Fermi and betatron mechanisms to electron acceleration become less and less important. When the guide field is sufficiently large, the contributions of the Fermi and betatron mechanisms are almost negligible.

79 ASTRONOMY AND ASTROPHYSICS↗

Fluid Dynamics Effects on Microstructure Prediction in Single Laser Tracks for Additive Manufacturing

The Laser Powder Bed Fusion Additive Manufacturing (LPBFAM) is one of the most important processes for the production of lightweight, cost-effective, complex, and high-performance enduse parts. At present, the cost and time associated with LPBFAM process development is very high due to a lack of fundamental process understanding. In this project, a multi-physics model was developed on a highly parallel open-source code, Truchas, with the ultimate goal of providing experimentally validated process maps for tailoring microstructure to achieve desired performance for LPBFAM. As a critical step towards fully LPBFAM modeling, modeling of single-track laser fusion (STLF) were conducted. Multi-physics simulations were conducted using Truchas for STLF by considering heat transfer, phase-change, fluid dynamics, surface tension phenomena, and evaporation. In order to assess the effect of the fluid dynamics on the solidification and ensuing microstructure, a heat transfer-and-solidification-only (HTS) model and a fully coupled heat-transfer-solidification and fluid-dynamics (HTS-FD) model were considered. In the HTS_FD model, the fluid flow was considered to be laminar while the molten alloy surface is assumed to be flat and non-deformable. This study is one the first attempt to understand the effect of the fluid flow on microstructure in STLF and LPBFAM. The results show that the fluid flow affects the solidification and ensuing microstructure in STLF. In order to validate Truchas for STLF and LPBFAM modeling, experimental data, which was obtained at GE Global Research (GEGR) for liquid pool shape and microstructure, were compared with those from numerical simulation results. For the keyhole regime case, numerical simulation results indicate that the melt-pool shape was typical to that of the conduction case with very large deviations from the measured melt-pool depths. Two sensitivity studies were conducted by varying the evaporation flux and the surface tension coefficient at fixed laser absorptivities of 0.5 and 0.3, respectively. The minimum value for the overall combined error, between the calculated values with the HTS_FD model and measured values for both the width and height of the melt-pool, was attained for a surface tension coefficient of -0.75e-4 N/m. An analytical solidification model for the columnar-to-equiaxed transition (CET) in rapid solidification was used to assess the microstructure variation within the melt-pool. From numerical simulation results, thermal gradient, G, and solidification velocity, V, were obtained in order to predict the microstructure type (e.g., dendritic, cellular). For the fluid-flow HTS_FD model, G was found to exhibit a maximum at the extremity of the solidified pool (i.e., at the free surface). By contrast, for HTS simulations, G was found to exhibit a maximum around the entire edge of the solidified pool. For the HTS_FD simulations, the minimum values of the cooling rate, GV, were found to be approximately half than their corresponding values for HTS simulations. By contrast with the min GV values, the maximum values for GV were found to occur for the HTS_FD simulations. Thus, HTS_FD simulations were found to exhibit a wider range of cooling rates than the HTS simulations. Concerning the solidification maps, the variation of the solidification velocity, V, as function of the thermal gradient, G, was obtained. It was found that the fluid flow model results (HTS_FD) exhibited an increased spread in the V(G) variation within the melt pool with respect to the HTS model results (without the fluid flow). A preliminary correlation model for the primary dendrite arm spacing (PDAS) based on a power law dependence on thermal gradient and solidification velocity was used to estimate PDAS from HTS and HTS_FD simulations. For low laser powers and low laser speeds, the PDAS obtained 6 with the fluid dynamics model (HTS_FD) was larger by more than 30% with respect to the PDAS calculated with the simple HTS model. In the second part of the project, fifty-seven simulations were conducted in order to obtain data on microstructure variables that can be used for process map development. In order to cover the entire processing space, thirty-three cases were selected STLF process simulations at six power levels and six laser scanning speeds. The 57 STLF process simulations were conducted as follows: 33 simulations with the heat-transfer-only (HTS) model and 24 simulations with the fluid flow model (HTS_FD). The solidification map data shows that for all simulations, a columnar dendritic microstructure would be expected. It was found that the minimum, average, and maximum thermal gradient exhibit exponential variations with respect to a process variable defined as the ratio between the power and square root of scan speed.

36 MATERIALS SCIENCE↗

Magnetic Reconnection in Solar Flares [Slides]

Magnetic reconnection is the underlying mechanism in solar flares. Magnetic reconnection is an astrophysical process in an electrically conducting plasma where neighboring magnetic field lines running anti-parallel break and reconnection therefore exchanging the direction of their paths. The result of this process is the magnetic field energy converting to kinetic and thermal energy which is explosively released. The scale of a solar is extremely large compared to Earth so kinetic simulations are unable to portray an accurate model. In this project, we are using an astrophysical magnetohydrodynamic (MHD) simulation code known as Athena++ with the reconnection problem generator file created by Li et al. 2018. To further study this simulation, we began testing a heating/cooling term. The results of the original file will be compared with the modified file with an implemented heating/cooling term to better understand how solar flares function.

79 ASTRONOMY AND ASTROPHYSICS↗

MAGNETIC RECONNECTION BETWEEN SMALL-SCALE LOOPS OBSERVED WITH THE NEW VACUUM SOLAR TELESCOPE

Using the high tempo-spatial resolution Hα images observed with the New Vacuum Solar Telescope, we report solid observational evidence of magnetic reconnection between two sets of small-scale, anti-parallel loops with an X-shaped topology. The reconnection process contains two steps: a slow step with a duration of more than several tens of minutes, and a rapid step lasting for only about three minutes. During the slow reconnection, two sets of anti-parallel loops gradually reconnect, and new loops are formed and stacked together. During the rapid reconnection, the anti-parallel loops approach each other quickly, and then rapid reconnection takes place, resulting in the disappearance of the former loops. In the meantime, new loops are formed and separate. The region between the approaching loops is brightened, and the thickness and length of this region are determined to be about 420 km and 1.4 Mm, respectively. During the rapid reconnection process, obvious brightenings at the reconnection site and apparent material ejections outward along reconnected loops are observed. These observed signatures are consistent with predictions by reconnection models. We suggest that the successive slow reconnection changes the conditions around the reconnection site and triggers instabilities, thus leading to the rapid approach of the anti-parallel loops and resulting in the rapid reconnection.

79 ASTRONOMY AND ASTROPHYSICS↗