Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “performance map”

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 1,495 records · Page 83
All sourcesNASA NTRSDOE OSTI
NASA NTRS1997

By Hand or Not By-Hand: A Case Study of Alternative Approaches to Parallelize CFD Applications

While parallel processing promises to speed up applications by several orders of magnitude, the performance achieved still depends upon several factors, including the multiprocessor architecture, system software, data distribution and alignment, as well as the methods used for partitioning the application and mapping its components onto the architecture. The existence of the Gorden Bell Prize given out at Supercomputing every year suggests that while good performance can be attained for real applications on general purpose multiprocessors, the large investment in man-power and time still has to be repeated for each application-machine combination. As applications and machine architectures become more complex, the cost and time-delays for obtaining performance by hand will become prohibitive. Computer users today can turn to three possible avenues for help: parallel libraries, parallel languages and compilers, interactive parallelization tools. The success of these methodologies, in turn, depends on proper application of data dependency analysis, program structure recognition and transformation, performance prediction as well as exploitation of user supplied knowledge. NASA has been developing multidisciplinary applications on highly parallel architectures under the High Performance Computing and Communications Program. Over the past six years, the transition of underlying hardware and system software have forced the scientists to spend a large effort to migrate and recede their applications. Various attempts to exploit software tools to automate the parallelization process have not produced favorable results. In this paper, we report our most recent experience with CAPTOOL, a package developed at Greenwich University. We have chosen CAPTOOL for three reasons: 1. CAPTOOL accepts a FORTRAN 77 program as input. This suggests its potential applicability to a large collection of legacy codes currently in use. 2. CAPTOOL employs domain decomposition to obtain parallelism. Although the fact that not all kinds of parallelism are handled may seem unappealing, many NASA applications in computational aerosciences as well as earth and space sciences are amenable to domain decomposition. 3. CAPTOOL generates code for a large variety of environments employed across NASA centers: MPI/PVM on network of workstations to the IBS/SP2 and CRAY/T3D.

Yan, Jerry C.↗
DOE OSTI2024

ICED: An Integrated CGRA Framework Enabling DFVS-Aware Acceleration

oarse-grained reconfigurable arrays (CGRAs) are a promising solution to enable energy-efficient acceleration of applications from different domains. By leveraging reconfiguration at the functional level, they can adapt to significantly different computational patterns. Existing CGRA mapping approaches extract instruction-level parallelism, exploit loop-pipelining opportunities, guarantee the data dependency, and target high throughput of a given loop. However, the recurrence data-dependency in the DFG and the mismatch between required and available computing/communication resources complicate the mapping, and might lead to significant unbalances in the utilization of the CGRA's tiles. This results in wasted power for tiles with low utilization. Applying dynamic voltage and frequency scaling (DVFS) can potentially solve this challenge and improve energy efficiency by adjusting voltage and frequency of different tiles independently. CGRAs have also been successful in accelerating data-dependent streaming applications. However, in these applications, the execution time of each kernel in the pipeline might dynamically vary depending on the characteristics of the input. This also leads to under-utilization of resources for the dynamically changing kernels that do not limit the application throughput. DVFS can also improve energy efficiency for these applications by dynamically changing the voltage and frequency levels of tiles that host non performance-constraining kernels. This paper proposes ICEDTEA -- an integrated DVFS-aware framework to map applications on CGRAs that support power islands. ICEDTEA proposes a CGRA architecture supporting DVFS islands at varying granularity (from a single tile to a group of tiles) and the related DVFS-aware compilation and mapping toolchain. ICEDTEA is the first work that introduces DVFS support for spatio-temporal CGRAs at power-island levels. The experimental evaluation shows that ICEDTEA improves average utilization by 2.3$\times$ and energy-efficiency by 1.32$\times$ over a conventional CGRA. With streaming applications, ICEDTEA improves energy efficiency by 1.12$\times$ over a state-of-the-art CGRA that introduces partial dynamic reconfiguration to adapt to variations in kernels' throughput.

Tan, Cheng↗
NASA NTRS1998

Development and Testing of Automatically Generated ACS Flight Software for the MAP Spacecraft

By integrating the attitude determination and control system (ACS) analysis and design, flight software development, and flight software testing processes, it is possible to improve the overall spacecraft development cycle, as well as allow for more thorough software testing. One of the ways to achieve this integration is to use code-generation tools to automatically generate components of the ACS flight software directly from a high-fidelity (HiFi) simulation. In the development of the Microwave Anisotropy Probe (MAP) spacecraft, currently underway at the NASA Goddard Space Flight Center, approximately 1/3 of the ACS flight software was automatically generated. In this paper, we will examine each phase of the ACS subsystem and flight software design life cycle: analysis, design, and testing. In the analysis phase, we scoped how much software we would automatically generate and created the initial interface. The design phase included parallel development of the HiFi simulation and the hand-coded flight software components. Everything came together in the test phase, in which the flight software was tested, using results from the HiFi simulation as one of the bases of comparison for testing. Because parts of the spacecraft HiFi simulation were converted into flight software, more care needed to be put into its development and configuration control to support both the HiFi simulation and flight software. The components of the HiFi simulation from which code was generated needed to be designed based on the fact that they would become flight software. This process involved such considerations as protecting against mathematical exceptions, using acceptable module and parameter naming conventions, and using an input/output interface compatible with the rest of the flight software. Maintaining good configuration control was an issue for the HiFi simulation and the flight software, and a way to track the two systems was devised. Finally, an integrated test approach was devised to support flight software testing at both the unit- and build-test levels using the HiFi simulation to generate data for performance verification. Another benefit of the simulation and code-generation application used on the MAP project is that it supported bringing flight software and test data into the HiFi simulation environment. It was possible to integrate parts of the hand-coded flight software into the HiFi simulation, and also possible to import flight software test data for comparison and performance verification. This capability was used to incorporate the flight software Kalman filter into the HiFi simulation. This enabled us to greatly increase the amount of testing that could be done on the filter, because we could exert a greater degree of control over the software-only simulation than over the flight software test environment. Also, since the simulation could be used to run the Kalman filter faster than real time, our testing efficiency was greatly increased. We will conclude our discussion with a summary of the lessons learned thus far using automatically- generated code for the MAP project, and the spacecraft status as we work towards our scheduled launch in the year 2000.

ODonnell, James R., Jr.↗
DOE OSTI2021

Case Study of Using Kokkos and SYCLs Performance-Portable Frameworks for Milc-Dslash Benchmark on NVIDIA, AMD and Intel GPUs

Six of the top ten supercomputers in the TOP500 list from June 2021 rely on NVIDIA GPUs to achieve their peak compute bandwidth. With the announcement of Aurora, Frontier, and El Capitan, Intel and AMD have also entered the domain of providing GPUs for scientific computing. A consequence of the increased diversity in the GPU landscape is the emergence of portable programming models such as Kokkos, SYCL, OpenCL, and OpenMP, which allow application developers to maintain a single-source code across a diverse range of hardware architectures. While the portable frameworks try to optimize the compute resource usage on a given architecture, it is the programmers responsibility to expose parallelism in an application that can take advantage of thousands of processing elements available on GPUs. In this paper, we introduce a GPU-friendly parallel implementation of Milc-Dslash that exposes multiple hierarchies of parallelism in the algorithm. Milc-Dslash was designed to serve as a benchmark with highly optimized matrix-vector multiplications to measure the resource utilization on the GPU systems. The parallel hierarchies in the Milc-Dslash algorithm are mapped onto a target hardware using Kokkos and SYCL programming models. We present the performance achieved by Kokkos and SYCL implementations of Milc-Dslash on NVIDIA A100 GPU, AMD MI100 GPU, and Intel Gen9 GPU. Additionally, we compare the Kokkos and SYCL performances with those obtained from the versions written in CUDA and HIP programming models on NVIDIA A100 GPU and AMD MI100 GPU, respectively.

Dufek, Amanda S↗
NASA NTRS1999

Preliminary Correlations of Gravity and Topography from Mars Global Surveyor

The Mars Global Surveyor (MGS) spacecraft is currently in a 400-km altitude polar mapping orbit and scheduled to begin global mapping of Mars in March of 1999. Doppler tracking data collected in this Gravity Calibration Orbit prior to the nominal mapping mission combined with observations from the MGS Science Phasing Orbit in Spring - Summer 1999 and the Viking and mariner 9 orbiters has led to preliminary high resolution gravity fields. Spherical harmonic expansions have been performed to degree and order 70 and are characterized by the first high spatial resolution coverage of high latitudes. Topographic mapping by the Mars Orbiter Laser Altimeter on MGS is providing measurements of the height of the martian surface with sub-meter vertical resolution and 5-30 m absolute accuracy. Data obtained during the circular mapping phase are expected to provide the first high resolution measurements of surface heights in the southern hemisphere. The combination of gravity and topography measurements provides information on the structure of the planetary interior, i.e. the rigidity and distribution of internal density. The observations can also be used to address the mechanisms of support of surface topography. Preliminary results of correlations of gravity and topography at long planetary wavelengths will be presented and the implications for internal structure will be addressed.

Zuber, M. T.↗
DOE OSTI2024

ATLAS-MAP: An Automated Test Station for Gated Electronic Transport Measurements

The diversification of electronic materials in devices provides a strong incentive for methods to rapidly correlate device performance with fabrication decisions. In this work, we present a low-cost automated test station for gated electronic transport measurements of field-effect transistors. Utilizing open-source PyMeasure libraries for transparent instrument control, the “ATLAS-MAP” system serves as a customizable interface between sourcemeters and samples under test and is programmed to conduct transfer curve and van der Pauw methods with static and sweeping gate voltages. Zinc oxide transistors of variable thickness (5, 10, and 20 nm) and channel size (50 μm to 3 mm, of equal length and width) were fabricated to validate the design. Standardization of testing procedures and raw data formatting enabled automated data analysis. A detailed list of parts and code files for the system are provided.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Analysis of Interpretable Data Representations for 4D-STEM Using Unsupervised Learning

Abstract Understanding the structure of materials is crucial for engineering devices and materials with enhanced performance. Four-dimensional scanning transmission electron microscopy (4D-STEM) is capable of mapping nanometer-scale local crystallographic structure over micron-scale field of views. However, 4D-STEM datasets can contain tens of thousands of images from a wide variety of material structures, making it difficult to automate detection and classification of structures. Traditional automated analysis pipelines for 4D-STEM focus on supervised approaches, which require prior knowledge of the material structure and cannot describe anomalous or deviant structures. In this article, a pipeline for engineering 4D-STEM feature representations for unsupervised clustering using non-negative matrix factorization (NMF) is introduced. Each feature is evaluated using NMF and results are presented for both simulated and experimental data. It is shown that some data representations more reliably identify overlapping grains. Additionally, real space refinement is applied to identify spatially distinct sample regions, allowing for size and shape analysis to be performed. This work lays the foundation for improved analysis of nanoscale structural features in materials that deviate from expected crystallographic arrangement using 4D-STEM.

Bruefach, Alexandra (ORCID:0000000209323477)↗
NASA NTRS1991

Magellan battery operations: An overview

The Magellan Spacecraft's mission to map Venus's surface provides a unique application for Nickel-Cadmium batteries. An overview of the spacecraft, power subsystem, battery, and mission requirements are presented. The reliability and performance of the batteries were extensively studied with a comprehensive test and analysis program. Actual data for cruise and a limited number of mapping orbits is also presented.

Timmerman, Paul J.↗
DOE OSTI2023

The Kokkos OpenMPTarget Backend: Implementation and Lessons Learned

As the supercomputing landscape diversifies, solutions such as Kokkos to write vendor agnostic applications and libraries have risen in popularity. Kokkos provides a programming model designed for performance portability, which allows developers to write a single source implementation that can run efficiently on various architectures. At its heart, Kokkos maps parallel algorithms to architecture and vendor specific backends written in lower level programming models such as CUDA and HIP. Another approach to writing vendor agnostic parallel code is using OpenMP’s directives based approach, which lets developers annotate code to express parallelism. It is implemented at the compiler level and is supported by all major high performance computing vendors, as well as the primary Open Source toolchains GNU and LLVM. Since its inception, Kokkos has used OpenMP to parallelize on CPU architectures. In this paper, we explore leveraging OpenMP for a GPU backend and discuss the challenges we encountered when mapping the Kokkos APIs and semantics to OpenMP target constructs. As an exemplar workload we chose a simple conjugate gradient solver for sparse matrices. We find that performance on NVIDIA and AMD GPUs varies widely based on details of the implementation strategy and the chosen compiler. Furthermore, the performance of the OpenMP implementations decreases with increasing complexity of the investigated algorithms.

Gayatri, Rahulkumar↗
DOE OSTI2021

Impact of Dry Cooler Air-Side Performance on a sCO2 Power Cycle for a CSP Application

Uncertainty around the design and control of the supercritical CO2 power cycle must be reduced before this technology can be implemented for large-scale grid support. To better understand the day-to-day performance of an sCO2 cycle, off-design performance calculations must be included for all power block components, and performance assumptions must be removed. This study has expanded the modeled scope to include the air-side performance for the dry cooler and has incorporated discretized heat transfer calculations for both streams through the pre-cooler to better predict off-design performance. This study considered a recompression Brayton cycle in a concentrating solar power application. The cycle model utilized fixed sCO2 turbomachinery maps for the main compressor, recompressor, and expander operating to supply approximately 10 MW gross at the design point. Fixed vendor-supplied fan curves were used to calculate the air-side performance of the dry cooler. The primary heater was modeled considering both the sCO2 and heat transfer fluid streams. Off-design performance was predicted for an ambient temperature range of 0-55℃, a HTF temperature range of 705-735℃, and a HTF mass flow range of 50-105% of the design point value. To understand the importance of modeling the air-side performance, the cycle off-design performance was also calculated using a constant CO2 outlet temperature assumption and a constant approach temperature assumption for the dry cooler. Results show that using these assumptions can significantly alter the power output and cycle efficiency predictions.

14 SOLAR ENERGY↗
DOE OSTI2017

Consequences of spin-orbit coupling at the single hole level: Spin-flip tunneling and the anisotropic g factor

Hole transport experiments were performed on a gated double quantum dot device defined in a p-GaAs/AlGaAs heterostructure with a single hole occupancy in each dot. The charging diagram of the device was mapped out using charge detection confirming that the single hole limit is reached. In that limit, a detailed study of the two-hole spin system was performed using high bias magnetotransport spectroscopy. In contrast to electron systems, the hole spin was found not to be conserved during interdot resonant tunneling. This allows one to fully map out the two-hole energy spectrum as a function of the magnitude and the direction of the external magnetic field. As a result, the heavy-hole g factor was extracted and shown to be strongly anisotropic, with a value of 1.45 for a perpendicular field and close to zero for an in-plane field as required for hybridizing schemes between spin and photonic quantum platforms.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2017

Recall Performance for Content-Addressable Memory Using Adiabatic Quantum Optimization

A content-addressable memory (CAM) stores key-value associations such that the key is recalled by providing its associated value. While CAM recall is traditionally performed using recurrent neural network models, we show how to solve this problem using adiabatic quantum optimization. Our approach maps the recurrent neural network to a commercially available quantum processing unit by taking advantage of the common underlying Ising spin model. We then assess the accuracy of the quantum processor to store key-value associations by quantifying recall performance against an ensemble of problem sets. We observe that different learning rules from the neural network community influence recall accuracy but performance appears to be limited by potential noise in the processor. The strong connection established between quantum processors and neural network problems supports the growing intersection of these two ideas.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1998

MAP stability, design, and analysis

The Microwave Anisotropy Probe (MAP) is a follow-on to the Differential Microwave Radiometer (DMR) instrument on the Cosmic Background Explorer (COBE) spacecraft. The design and analysis of the MAP attitude control system (ACS) have been refined since work previously reported. The full spacecraft and instrument flexible model was developed in NASTRAN, and the resulting flexible modes were plotted and reduced with the Modal Significance Analysis Package (MSAP). The reduced-order model was used to perform the linear stability analysis for each control mode, the results of which are presented in this paper. Although MAP is going to a relatively disturbance-free Lissajous orbit around the Earth-Sun L(2) Lagrange point, a detailed disturbance-torque analysis is required because there are only a small number of opportunities for momentum unloading each year. Environmental torques, including solar pressure at L(2), aerodynamic and gravity gradient during phasing-loop orbits, were calculated and simulated. Thruster plume impingement torques that could affect the performance of the thruster modes were estimated and simulated, and a simple model of fuel slosh was derived to model its effect on the motion of the spacecraft. In addition, a thruster mode linear impulse controller was developed to meet the accuracy requirements of the phasing loop burns. A dynamic attitude error limiter was added to improve the performance of the ACS during large attitude slews. The result of this analysis is a stable ACS subsystem that meets all of the mission's requirements.

Ericsson-Jackson, A. J.↗
DOE OSTI2021

Fast Grain Mapping with Sub-Nanometer Resolution Using 4D-STEM with Grain Classification by Principal Component Analysis and Non-Negative Matrix Factorization

High-throughput grain mapping with sub-nanometer spatial resolution is demonstrated using scanning nanobeam electron diffraction (also known as 4D scanning transmission electron microscopy, or 4D-STEM) combined with high-speed direct-electron detection. An electron probe size down to 0.5 nm in diameter is used and the sample investigated is a gold–palladium nanoparticle catalyst. Computational analysis of the 4D-STEM data sets is performed using a disk registration algorithm to identify the diffraction peaks followed by feature learning to map the individual grains. Two unsupervised feature learning techniques are compared: principal component analysis (PCA) and non-negative matrix factorization (NNMF). The characteristics of the PCA versus NNMF output are compared and the potential of the 4D-STEM approach for statistical analysis of grain orientations at high spatial resolution is discussed.

47 OTHER INSTRUMENTATION↗
DOE OSTI2021

Assimilation of citizen science data in snowpack modeling using a new snow data set: Community Snow Observations

A physically based snowpack evolution and redistribution model was used to test the effectiveness of assimilating crowd-sourced snow depth measurements collected by citizen scientists. The Community Snow Observations project gathers, stores, and distributes measurements of snow depth recorded by recreational users and snow professionals in high mountain environments. These citizen science measurements are valuable since they come from terrain that is relatively undersampled and can offer in situ snow information in locations where snow information is sparse or nonexistent. The present study investigates (1) the improvements to model performance when citizen science measurements are assimilated, and (2) the number of measurements necessary to obtain those improvements. Model performance is assessed by comparing time series of observed (snow pillow) and modeled snow water equivalent values, by comparing spatially distributed maps of observed (remotely sensed) and modeled snow depth, and by comparing fieldwork results from within the study area. The results demonstrate that few citizen science measurements are needed to obtain improvements in model performance, and these improvements are found in 62 % to 78 % of the ensemble simulations, depending on the model year. Model estimations of total water volume from a subregion of the study area also demonstrate improvements in accuracy after CSO measurements have been assimilated. These results suggest that even modest measurement efforts by citizen scientists have the potential to improve efforts to model snowpack processes in high mountain environments, with implications for water resource management and process-based snow modeling.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1990

Solving unstructured grid problems on massively parallel computers

A highly parallel graph mapping technique that enables one to efficiently solve unstructured grid problems on massively parallel computers is presented. Many implicit and explicit methods for solving discretized partial differential equations require each point in the discretization to exchange data with its neighboring points every time step or iteration. The cost of this communication can negate the high performance promised by massively parallel computing. To eliminate this bottleneck, the graph of the irregular problem is mapped into the graph representing the interconnection topology of the computer such that the sum of the distances that the messages travel is minimized. It is shown that using the heuristic mapping algorithm significantly reduces the communication time compared to a naive assignment of processes to processors.

Hammond, Steven W.↗
DOE OSTI2020

Ultrafast structural response to charge redistribution within a photosinthetic reaction centre

This data set contains stream files, difference maps and SVD maps obtained from data processing of time-resolved serial femtosecond crystallography measurements performed at LCLS XFEL, used to construct the images and perform the analysis published in the paper "Ultrafast structural response to charge redistribution within a photosinthetic reaction centre". Two (2) time points plus a dark state from 2015 ('a' label) as well as five (5) time points plus a dark state from 2016 ('b' label) are provided.

CXI↗