Search NASA⌕ Search

SEARCH · Search NASA

Results for “high performance and distributed computing”

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 613 records · Page 34
All sourcesNASA NTRSDOE OSTI
NASA NTRS1986

Optical processing for future computer networks

In the development of future data management systems, such as the NASA Space Station, a major problem represents the design and implementation of a high performance communication network which is self-correcting and repairing, flexible, and evolvable. To obtain the goal of designing such a network, it will be essential to incorporate distributed adaptive network control techniques. The present paper provides an outline of the functional and communication network requirements for the Space Station data management system. Attention is given to the mathematical representation of the operations being carried out to provide the required functionality at each layer of communication protocol on the model. The possible implementation of specific communication functions in optics is also considered.

Husain, A.↗
DOE OSTI2017

Thermal-hydraulics modeling for prototype testing of the W7-X high heat flux scraper element

The long-pulse operation of the Wendelstein 7-X (W7-X) stellarator experiment is scheduled to begin in 2020. This operational phase will be equipped with water-cooled plasma facing components to allow for longer pulse durations. Certain simulated plasma scenarios have been shown to produce heat fluxes that surpass the technological limits on the edges of the divertor target elements during steady-state operation. In order to reduce the heat load on the target elements, the addition of a “scraper element” (SE) is under investigation. The SE is composed of 24 water-cooled carbon fiber reinforced carbon composite monoblock units. Multiple full-scale prototypes have been tested in the GLADIS high heat flux test facility. Previous computational studies revealed discrepancies between the simulations and experimental measurements. In this work, single-phase thermal-hydraulics modeling was performed in ANSYS CFX to identify potential causes for such discrepancies. Possible explanations investigated were the effects of a non-uniform thermal contact resistance and a potential misalignment of the monoblock fibers. And while the difference between the experimental and computational results was not resolved by a non-uniform thermal contact resistance, the computational results provided insight into the potential performance of a W7-X monoblock unit. Circumferential temperature distributions highlighted the expected boiling regions of such a unit. Finally, simulations revealed that modest angles of fiber misalignment in the monoblocks result in asymmetries at the unit edges and provide temperature differences similar to the experimental results.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1999

Engineering Overview of a Multidisciplinary HSCT Design Framework Using Medium-Fidelity Analysis Codes

An objective of the HPCC Program at NASA Langley has been to promote the use of advanced computing techniques to more rapidly solve the problem of multidisciplinary optimization of a supersonic transport configuration. As a result, a software system has been designed and is being implemented to integrate a set of existing discipline analysis codes, some of them CPU-intensive, into a distributed computational framework for the design of a High Speed Civil Transport (HSCT) configuration. The proposed paper will describe the engineering aspects of integrating these analysis codes and additional interface codes into an automated design system. The objective of the design problem is to optimize the aircraft weight for given mission conditions, range, and payload requirements, subject to aerodynamic, structural, and performance constraints. The design variables include both thicknesses of structural elements and geometric parameters that define the external aircraft shape. An optimization model has been adopted that uses the multidisciplinary analysis results and the derivatives of the solution with respect to the design variables to formulate a linearized model that provides input to the CONMIN optimization code, which outputs new values for the design variables. The analysis process begins by deriving the updated geometries and grids from the baseline geometries and grids using the new values for the design variables. This free-form deformation approach provides internal FEM (finite element method) grids that are consistent with aerodynamic surface grids. The next step involves using the derived FEM and section properties in a weights process to calculate detailed weights and the center of gravity location for specified flight conditions. The weights process computes the as-built weight, weight distribution, and weight sensitivities for given aircraft configurations at various mass cases. Currently, two mass cases are considered: cruise and gross take-off weight (GTOW). Weights information is obtained from correlations of data from three sources: 1) as-built initial structural and non-structural weights from an existing database, 2) theoretical FEM structural weights and sensitivities from Genesis, and 3) empirical as-built weight increments, non-structural weights, and weight sensitivities from FLOPS. For the aeroelastic analysis, a variable-fidelity aerodynamic analysis has been adopted. This approach uses infrequent CPU-intensive non-linear CFD to calculate a non-linear correction relative to a linear aero calculation for the same aerodynamic surface at an angle of attack that results in the same configuration lift. For efficiency, this nonlinear correction is applied after each subsequent linear aero solution during the iterations between the aerodynamic and structural analyses. Convergence is achieved when the vehicle shape being used for the aerodynamic calculations is consistent with the structural deformations caused by the aerodynamic loads. To make the structural analyses more efficient, a linearized structural deformation model has been adopted, in which a single stiffness matrix can be used to solve for the deformations under all the load conditions. Using the converged aerodynamic loads, a final set of structural analyses are performed to determine the stress distributions and the buckling conditions for constraint calculation. Performance constraints are obtained by running FLOPS using drag polars that are computed using results from non-linear corrections to the linear aero code plus several codes to provide drag increments due to skin friction, wave drag, and other miscellaneous drag contributions. The status of the integration effort will be presented in the proposed paper, and results will be provided that illustrate the degree of accuracy in the linearizations that have been employed.

Weston, R. P.↗
DOE OSTI2018

High-Performance Multi-Mode Ptychography Reconstruction on Distributed GPUs

Abstract—Ptychography is an emerging imaging technique that is able to provide wavelength-limited spatial resolution from spec-imen with extended lateral dimensions. As a scanning microscopy method, a typical two-dimensional image requires a number of data frames. As a diffraction-based imaging technique, the real-space image has to be recovered through iterative reconstruction algorithms. Due to these two inherent aspects, a ptychographic reconstruction is generally a computation-intensive and time-consuming process, which limits the throughput of this method. We report an accelerated version of the multi-mode difference map algorithm for ptychography reconstruction using multiple distributed GPUs. This approach leverages available scientific computing packages in Python, including mpi4py and PyCUDA, with the core computation functions implemented in CUDA C. We find that interestingly even with MPI collective communications, the weak scaling in the number of GPU nodes can still remain nearly constant. Most importantly, for realistic diffraction measurements, we observe a speedup ranging from a factor of 10 to 103 depending on the data size, which reduces the reconstruction time remarkably from hours to typically about 1 minute and is thus critical for real-time data processing and visualization.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Multidimensional Vlasov–Poisson Simulations with High-order Monotonicity- and Positivity-preserving Schemes

We develop new numerical schemes for Vlasov–Poisson equations with high-order accuracy. Our methods are based on a spatially monotonicity-preserving (MP) scheme and are modified suitably so that the positivity of the distribution function is also preserved. We adopt an efficient semi-Lagrangian time integration scheme that is more accurate and computationally less expensive than the three-stage TVD Runge–Kutta integration. We apply our spatially fifth- and seventh-order schemes to a suite of simulations of collisionless self-gravitating systems and electrostatic plasma simulations, including linear and nonlinear Landau damping in one dimension and Vlasov–Poisson simulations in a six-dimensional phase space. The high-order schemes achieve a significantly improved accuracy in comparison with the third-order positive-flux-conserved scheme adopted in our previous study. With the semi-Lagrangian time integration, the computational cost of our high-order schemes does not significantly increase, but remains roughly the same as that of the third-order scheme. Vlasov–Poisson simulations on 128{sup 3}×128{sup 3} mesh grids have been successfully performed on a massively parallel computer.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

Tailoring the Properties of a Ni-Based Superalloy via Modification of the Forging Process: an ICME Approach to Fatigue Performance

Traditionally, material design and property modifications are usually associated with compositional changes. However, subtle changes in the manufacturing process parameters can also have a dramatic effect on the resulting material properties. In this work, an integrated computational materials engineering (ICME) framework is adopted to tailor the fatigue performance of a Ni-based superalloy, RR1000. An existing fatigue model is used to identify microstructural features that promote enhanced fatigue life, namely a uniform, fine grain size distribution, random orientation, a distinct grain boundary distribution (specifically high twin boundary density and limited low-angle grain boundaries). A deformation mechanism map and process models for grain boundary engineering of RR1000 are used to identify the optimal thermo-mechanical processing parameters to realize these desirable microstructural features. For validation, small-scale forgings of RR1000 were produced and heat-treated to attain fine grain and coarse grain microstructures that represent the conventionally processed and grain boundary engineered (GBE) conditions, respectively. For each of the four microstructural variants of RR1000, the twin density and grain size were characterized and were in agreement with the desired microstructural attributes. In order to validate the deformation mechanisms and fatigue behavior of the material, high-resolution digital image correlation was performed to generate strain maps relative to the microstructural features. The high density of twin boundaries was confirmed to inhibit the length of slip bands, which is directly attributed to extended fatigue life. Thus, this study demonstrated the successful role of models, both process and performance, in the design and manufacture of Ni-based superalloy disk forgings.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Multigrid treatment of implicit continuum diffusion

Implicit treatment of diffusive terms of various differential orders common in continuum mechanics modeling, such as computational fluid dynamics, is investigated with spectral and multigrid algorithms in non-periodic 2D domains. In doubly periodic time dependent problems these terms are handled efficiently by spectral methods, but in non-periodic systems solved with distributed memory parallel computing and 2D domain decomposition, this efficiency is lost for a large number of processors. Here, we built and present here a multigrid algorithm for these types of problems that outperforms a spectral solution employing the highly optimized FFTW library. This solver is suitable for high performance computing and may be able to efficiently treat implicit diffusion of arbitrary order by introducing auxiliary equations of lower order. We test these solvers for fourth and sixth order diffusion with harmonic test functions as well as turbulent 2D magnetohydrodynamic simulations. It is also shown that an anisotropic operator without mixed-derivative terms improves model accuracy and speed, and we examine the impact that the various diffusion operators have on the energy, the enstrophy, and the qualitative aspect of a simulation.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRS1990

Local-Area-Network Simulator

Local Area Network Extensible Simulator (LANES) computer program provides method for simulating performance of high-speed local-area-network (LAN) technology. Developed as design and analysis software tool for networking computers on board proposed Space Station. Load, network, link, and physical layers of layered network architecture all modeled. Mathematically models according to different lower-layer protocols: Fiber Distributed Data Interface (FDDI) and Star*Bus. Written in FORTRAN 77.

Gibson, Jim↗
NASA NTRS1990

Advanced sensors and instrumentation

NASA is currently investigating the readiness of Advanced Sensors and Instrumentation to meet the requirements of new initiatives in space. The following technical objectives and technologies are briefly discussed: smart and nonintrusive sensors; onboard signal and data processing; high capacity and rate adaptive data acquisition systems; onboard computing; high capacity and rate onboard storage; efficient onboard data distribution; high capacity telemetry; ground and flight test support instrumentation; power distribution; and workstations, video/lighting. The requirements for high fidelity data (accuracy, frequency, quantity, spatial resolution) in hostile environments will continue to push the technology developers and users to extend the performance of their products and to develop new generations.

Calloway, Raymond S.↗
NASA NTRS2002

A Facility and Architecture for Autonomy Research

Autonomy is a key enabling factor in the advancement of the remote robotic exploration. There is currently a large gap between autonomy software at the research level and software that is ready for insertion into near-term space missions. The Mission Simulation Facility (MST) will bridge this gap by providing a simulation framework and suite of simulation tools to support research in autonomy for remote exploration. This system will allow developers of autonomy software to test their models in a high-fidelity simulation and evaluate their system's performance against a set of integrated, standardized simulations. The Mission Simulation ToolKit (MST) uses a distributed architecture with a communication layer that is built on top of the standardized High Level Architecture (HLA). This architecture enables the use of existing high fidelity models, allows mixing simulation components from various computing platforms and enforces the use of a standardized high-level interface among components. The components needed to achieve a realistic simulation can be grouped into four categories: environment generation (terrain, environmental features), robotic platform behavior (robot dynamics), instrument models (camera/spectrometer/etc.), and data analysis. The MST will provide basic components in these areas but allows users to plug-in easily any refined model by means of a communication protocol. Finally, a description file defines the robot and environment parameters for easy configuration and ensures that all the simulation models share the same information.

Pisanich, Greg↗
DOE OSTI2017

NPTFit: A Code Package for Non-Poissonian Template Fitting

We present NPTFit, an open-source code package, written in Python and Cython, for performing non-Poissonian template fits (NPTFs). The NPTF is a recently developed statistical procedure for characterizing the contribution of unresolved point sources (PSs) to astrophysical data sets. The NPTF was first applied to Fermi gamma-ray data to provide evidence that the excess of ∼GeV gamma-rays observed in the inner regions of the Milky Way likely arises from a population of sub-threshold point sources, and the NPTF has since found additional applications studying sub-threshold extragalactic sources at high Galactic latitudes. The NPTF generalizes traditional astrophysical template fits to allow for the ability to search for populations of unresolved PSs that may follow a given spatial distribution. NPTFit builds upon the framework of the fluctuation analyses developed in X-ray astronomy, thus it likely has applications beyond those demonstrated with gamma-ray data. The NPTFit package utilizes novel computational methods to perform the NPTF efficiently. The code is available at http://github.com/bsafdi/NPTFit and up-to-date and extensive documentation may be found at http://nptfit.readthedocs.io.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2020

Modular performance prediction for scientific workflows using Machine Learning

Scientific workflows provide an opportunity for declarative computational experiment design in an intuitive and efficient way. A distributed workflow is typically executed on a variety of resources, and it uses a variety of computational algorithms or tools to achieve the desired outcomes. Such a variety imposes additional complexity in scheduling these workflows on large scale computers. As computation becomes more distributed, insights into expected workload that a workflow presents become critical for effective resource allocation. In this paper, we present a modular framework that leverages Machine Learning for creating precise performance predictions of a workflow. The central idea is to partition a workflow in such a way that makes the task of forecasting each atomic unit manageable and gives us a way to combine the individual predictions efficiently. We recognize a combination of an executable and a specific physical resource as a single module. This gives us a handle to characterize workload and machine power as a single unit of prediction. Overall, our modular technique of creating atomic modules and deployment of longest-path approach to estimate workflow performance, allows the framework to adapt to highly complex nested directed acyclic workflows and scale to new scenarios, since it does not make assumptions of underlying workflow structure. We present performance estimation results of independent workflow modules executed on the XSEDE SDSC Comet cluster using various Machine Learning algorithms. The results provide insights into the behavior and effectiveness of different algorithms in the context of scientific workflow performance prediction.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Parallel streaming between heterogeneous HPC resources for real-time analysis

Performing analysis or generating visualizations concurrently with high performance simulations can yield great benefits compared to post-processing data. Writing and reading large volumes of data can be reduced or eliminated, thereby producing an I/O cost savings. One such method for concurrent simulation and analysis is in transit - streaming data from the resource running the simulation to a separate resource running the analysis. In transit analysis can be beneficial since computational resources may not have certain resources needed for visualization and analysis (e.g. GPUs) and to reduce the impact of performing analysis tasks to the run time of the simulation. When sending and receiving data in transit, data redistribution mechanisms are needed in order to support heterogeneous data layouts that may be required by the simulation and analysis applications. The work described in this paper compares two mechanisms for on-the-fly data redistribution when streaming data in parallel between two distributed memory applications. Finally, our results show that it is often more advantageous to stream data in the same layout as the sender and redistribute data amongst processes on the receiving end than to stream data in the final layout needed by the receiver.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2002

Software Engineering Support of the Third Round of Scientific Grand Challenge Investigations: Earth System Modeling Software Framework Survey

One of the most significant challenges in large-scale climate modeling, as well as in high-performance computing in other scientific fields, is that of effectively integrating many software models from multiple contributors. A software framework facilitates the integration task, both in the development and runtime stages of the simulation. Effective software frameworks reduce the programming burden for the investigators, freeing them to focus more on the science and less on the parallel communication implementation. while maintaining high performance across numerous supercomputer and workstation architectures. This document surveys numerous software frameworks for potential use in Earth science modeling. Several frameworks are evaluated in depth, including Parallel Object-Oriented Methods and Applications (POOMA), Cactus (from (he relativistic physics community), Overture, Goddard Earth Modeling System (GEMS), the National Center for Atmospheric Research Flux Coupler, and UCLA/UCB Distributed Data Broker (DDB). Frameworks evaluated in less detail include ROOT, Parallel Application Workspace (PAWS), and Advanced Large-Scale Integrated Computational Environment (ALICE). A host of other frameworks and related tools are referenced in this context. The frameworks are evaluated individually and also compared with each other.

Talbot, Bryan↗
DOE OSTI2020

Energy storage emerging: A perspective from the Joint Center for Energy Storage Research

Energy storage is an integral part of modern society. A contemporary example is the lithium (Li)-ion battery, which enabled the launch of the personal electronics revolution in 1991 and the first commercial electric vehicles in 2010. Most recently, Li-ion batteries have expanded into the electricity grid to firm variable renewable generation, increasing the efficiency and effectiveness of transmission and distribution. Important applications continue to emerge including decarbonization of heavy-duty vehicles, rail, maritime shipping, and aviation and the growth of renewable electricity and storage on the grid. This perspective compares energy storage needs and priorities in 2010 with those now and those emerging over the next few decades. The diversity of demands for energy storage requires a diversity of purpose-built batteries designed to meet disparate applications. Advances in the frontier of battery research to achieve transformative performance spanning energy and power density, capacity, charge/discharge times, cost, lifetime, and safety are highlighted, along with strategic research refinements made by the Joint Center for Energy Storage Research (JCESR) and the broader community to accommodate the changing storage needs and priorities. Innovative experimental tools with higher spatial and temporal resolution, in situ and operando characterization, first-principles simulation, high throughput computation, machine learning, and artificial intelligence work collectively to reveal the origins of the electrochemical phenomena that enable new means of energy storage. This knowledge allows a constructionist approach to materials, chemistries, and architectures, where each atom or molecule plays a prescribed role in realizing batteries with unique performance profiles suitable for emergent demands.

Trahey, Lynn↗
NASA NTRS2021

Machine Learning Based Path Planning for Improved Rover Navigation

Enhanced AutoNav (ENav), the baseline surface navigation software for NASA’s Perseverance rover, sorts a list of candidate paths for the rover to traverse, then uses the Approximate Clearance Evaluation (ACE) algorithm to evaluate whether the most highly ranked paths are safe. ACE is crucial for maintaining the safety of the rover, but is computationally expensive. If the most promising candidates in the list of paths are all found to be infeasible, ENav must continue to search the list and run time-consuming ACE evaluations until a feasible path is found. In this paper, we present two heuristics that, given a terrain heightmap around the rover, produce cost estimates that more effectively rank the candidate paths before ACE evaluation. The first heuristic uses Sobel operators and convolution to incorporate the cost of traversing high-gradient terrain. The second heuristic uses a machine learning (ML) model to predict areas that will be deemed untraversable by ACE. We used physics simulations to collect training data for the ML model and to run Monte Carlo trials to quantify navigation performance across a variety of terrains with various slopes and rock distributions. Compared to ENav's baseline performance, integrating the heuristics can lead to a significant reduction in ACE evaluations and average computation time per planning cycle, increase path efficiency, and maintain or improve the rate of successful traverses. This strategy of targeting specific bottlenecks with ML while maintaining the original ACE safety checks provides an example of how ML can be infused into planetary science missions and other safety-critical software.

Yue, Yisong↗
DOE OSTI2022

Accelerated Coupled Monte Carlo-Thermal Hydraulic Calculations using a Hybrid GTF-Diffusion-based Prediction Block: First Results

Accurate predictions of spatial power and temperature distributions require the coupling of a neutron transport solver with a thermal-hydraulic (TH) feedback. Nowadays, Monte Carlo (MC) codes are widely coupled to TH solvers, typically via a Picard iteration (PI) method, due to the higher fidelity that such frameworks can produce. To speed up a PI, a prediction step can produce an improved initial guess for a source distribution and feed it to the MC code. Recent work [1, 2] investigated a prediction step that uses generalized transfer functions (GTFs) to predict the macroscopic cross sections’ variations following a perturbation in TH properties, such as coolant density. The previous method also relied on first order perturbation (FOP) theory to predict perturbed power profiles, rather than using an expensive MC iterate. The implemented FOP method relied on generating a fission matrix from which the forward and adjoint eigenmodes were extracted and later used to for power calculations. The generation of the fission matrix can introduce a significant computational overhead, therefore undermining the performance of the proposed hybrid technique when applied to high-dimensional problems, e.g., full core calculations. This work attempts to improve the GTF-FOP prediction step by replacing the FOP solver with a nodal diffusion solver, thus eliminating the need to calculate a fission matrix. The GTF-diffusion step was tested for various moderator density perturbations. In each case, the predicted power distribution showed good agreement with the reference case. The latter is attributed to the generally good prediction of most spatially distributed macroscopic cross sections, except the transport cross section, which will become the focus of future work.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

Machine Learning-enabled Scalable Performance Prediction of Scientific Codes

Hardware architectures become increasingly complex as the compute capabilities grow to exascale. Here, we present the Analytical Memory Model with Pipelines (AMMP) of the Performance Prediction Toolkit (PPT). PPT-AMMP takes high-level source code and hardware architecture parameters as input and predicts runtime of that code on the target hardware platform, which is defined in the input parameters. PPT-AMMP transforms the code to an (architecture-independent) intermediate representation, then (i) analyzes the basic block structure of the code, (ii) processes architecture-independent virtual memory access patterns that it uses to build memory reuse distance distribution models for each basic block, and (iii) runs detailed basic-block level simulations to determine hardware pipeline usage. PPT-AMMP uses machine learning and regression techniques to build the prediction models based on small instances of the input code, then integrates into a higher-order discrete-event simulation model of PPT running on Simian PDES engine. We validate PPT-AMMP on four standard computational physics benchmarks and present a use case of hardware parameter sensitivity analysis to identify bottleneck hardware resources on different code inputs. We further extend PPT-AMMP to predict the performance of a scientific application code, namely, the radiation transport mini-app SNAP. To this end, we analyze multi-variate regression models that accurately predict the reuse profiles and the basic block counts. We validate predicted SNAP runtimes against actual measured times.

97 MATHEMATICS AND COMPUTING↗