Search NASA⌕ Search

SEARCH · Search NASA

Results for “performance benchmarking”

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 1,153 records · Page 64
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

An automated and portable method for selecting an optimal GPU frequency

Power consumption poses a significant challenge in current and emerging graphics processing unit (GPU) enabled high-performance computing systems. In modern GPUs, dynamic voltage frequency scaling (DVFS) appears to be a reliable control to regulate power consumption and performance. However, the DVFS design space is large - hence, brute-force approaches are infeasible to select the optimal frequency. Furthermore, no single frequency can be universally optimal for applications with varying computational intensities. Thus, the application's complexity and the availability of a wide range of frequency settings are a challenge in selecting the optimal frequency configuration for a given GPU workload. To that end, this paper proposes a systematic approach that consists of three steps. The feature characterization study identifies the fine-grain GPU utilization metrics that influence the power consumption and execution time of a given workload. To understand the performance, power, and energy consumption behaviors of a workload across GPU's DVFS design space, we derived analytical power and performance models using the identified fine-grain features. Here, it is shown that the same set of GPU utilization metrics can estimate both the power consumption and execution time while being agnostic of changes to frequency and input sizes. Applying a power control with the single objective of reducing power may cause performance degradation, leading to more energy consumption. A multi-objective approach is proposed to select the optimal GPU DVFS configuration for a workload that reduces power consumption with negligible degradation in performance. The evaluation was conducted using SPEC ACCEL benchmarks and three real applications - NAMD LAMMPS, and LSTM on NVIDIA GV100, GA100, and AMD MI210 GPUs. On average, real applications showed 29.6% energy savings with a performance loss of 5.2% on GA100 and 22.6% energy savings with a performance loss of 4.7% on GV100. Moreover, the proposed models are portable to real applications, GPU architectures, and vendors, and require metric collection at only the default frequency rather than all supported DVFS configurations. Additionally, we conducted a comparison between our models and the GPU assembly instructions (PTX)-based static models. The results revealed a significant reduction in the average error rates, with a decrease from 19.7% to 3.1% for power models and from 29.4% to 5.2% for performance models.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

A discrete integral transform for rapid spectral synthesis

Accurate synthetic spectra that rely on large Line-By-Line (LBL)-databases are used in a wide range of applications such as high temperature combustion, atmospheric re-entry, planetary surveillance and laboratory plasmas. Conventionally synthetic spectra are calculated by computing a lineshape for every spectral line in the database and adding those together, which may take multiple hours for large databases. In this paper we propose a new approach for spectral synthesis based on an integral transform: the synthetic spectrum is calculated as the integral over the product of a Voigt profile and a newly proposed three-dimensional “lineshape distribution function”, which is a function of spectral position and Gaussian- & Lorentzian width coordinates. A fast discrete version of this transform based on the Fast Fourier Transform (FFT) is proposed, which improves performance compared to the conventional approach by several orders of magnitude while maintaining accuracy. Strategies that minimize the discretization error are discussed. A Python implementation of the method is compared against state-of-the-art spectral code RADIS, and is since adopted as RADIS's default synthesis method. The synthesis of a benchmark CO2 spectrum consisting of 1.8 M spectral lines and 200k spectral points took only 3.1 s using the proposed method (1011 lines × spectral points/s), a factor ~300 improvement over the state-of-the-art, with the relative improvement generally increasing for higher number of lines and/or number of spectral points. Finally, an experimental GPU-implementation of the method was also benchmarked, which demonstrated another 2~3 orders performance increase, achieving up to 5 ∙ 10 14 lines × spectral points/s.

42 ENGINEERING↗
DOE OSTI2022

3-regular three-XORSAT planted solutions benchmark of classical and quantum heuristic optimizers

With current semiconductor technology reaching its physical limits, special-purpose hardware has emerged as an option to tackle specific computing-intensive challenges. Optimization in the form of solving quadratic unconstrained binary optimization problems, or equivalently Ising spin glasses, has been the focus of several new dedicated hardware platforms. These platforms come in many different flavors, from highly-efficient hardware implementations on digital-logic of established algorithms to proposals of analog hardware implementing new algorithms. In this work, we use a mapping of a specific class of linear equations whose solutions can be found efficiently, to a hard constraint satisfaction problem (three-regular three-XORSAT, or an Ising spin glass) with a 'golf-course' shaped energy landscape, to benchmark several of these different approaches. We perform a scaling and prefactor analysis of the performance of Fujitsu's digital annealer unit (DAU), the D-Wave advantage quantum annealer, a virtual MemComputing machine, Toshiba's simulated bifurcation machine (SBM), the SATonGPU algorithm from Bernaschi et al, and our implementation of parallel tempering. We identify the SATonGPU and DAU as currently having the smallest scaling exponent for this benchmark, with SATonGPU having a small scaling advantage and in addition having by far the smallest prefactor thanks to its use of massive parallelism. Furthermore, our work provides an objective assessment and a snapshot of the promise and limitations of dedicated optimization hardware relative to a particular class of optimization problems.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2017

Implementation of CMFD Acceleration Scheme in PROTEUS-MOC

The PROTEUS-MOC code is a three-dimensional (3D) transport code, which is being developed at Argonne National Laboratory under the DOE-NE's Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. It provides a fully consistent 3D discretization of the transport equation by combining the two-dimensional (2D) method of characteristics (MOC) for the planar direction with the discontinuous Galerkin method for the treatment of the axial variable. PROTEUS-MOC uses the unstructured finite element method (FEM) and thus allows the explicit representation of any extruded geometries. It also has a capability of massively parallel computation in energy group, angle, plane, and space using a top-level GMRES linear algebra solver. For parallel computation in space, it uses an unstructured domain decomposition by adapting METIS. The computational burden for direct heterogeneous core calculations based on unstructured geometries is excessive for practical reactor problems, and hence it is essential to implement an efficient acceleration scheme in PROTEUS-MOC. Recently, the coarse mesh finite difference (CMFD) method, which was originally applied to the nodal diffusion methods, has been successfully applied to the MOC transport solvers, resulting in a remarkable reduction in the number of fission source iterations. In addition, a modified CMFD method has been developed to overcome the limitation of the conventional CMFD that is applicable only to neutronics solvers with a conservative discretization. This CMFD method was derived to be consistent with the FEM, which is based on a nonconservative discretization. In this method, the neutron balance in each coarse mesh is enforced by introducing a pseudo absorption cross section (PAXS), and the well-established alternating solution process of CMFD and transport calculations is employed to accelerate source convergence. This modified CMFD acceleration scheme was adopted in PROTEUS-MOC to accelerate the fission source iteration. Pin-cell and assembly based CMFD schemes have been implemented by defining individual pin cells or assemblies as a coarse mesh. For this, an algorithm to map the finite elements to coarse meshes was developed, and the routines were developed to determine the coupling coefficients and PAXS of the CMFD equations from the transport solution. Preliminary performance tests were carried out using the C5G7 benchmark problem. This paper presents the main features of this CMFD acceleration scheme and the preliminary test results.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2021

Flow boiling in geometrically modified microchannels

Microchannels are a promising solution for high-heat-flux thermal management scenarios, including high-performance microelectronics cooling and power electronics cooling. However, thermohydraulic instabilities result from the rapid vapor bubble formation. The prior literature has examined several methods, including constricted inlet microchannels, expanding microchannels, and auxiliary jetting microchannels, to mitigate the effect of these instabilities. Computational fluid dynamics and heat transfer (CFD/HT) modeling of the flow boiling phenomena in these microchannel configurations has seen limited examination, and one-to-one numerical comparisons of the different mitigation strategies have not been performed. In the present investigation, CFD/HT analyses using a three-dimensional (3D) volume of fluid model coupled with a phase-change model for the interfacial heat and mass transfer were performed for multiple microchannel configurations (constricted inlet, expanding, and auxiliary jetting microchannels). A benchmark case of a rectangular microchannel was examined to quantify baseline thermohydraulic performance. Results demonstrated slight to significant thermal performance improvements for all cases, and significant pressure benefits for the expanding and jetting cases, consistent with experimental results in the literature. Bubble dynamics and visualization for the baseline and alternative configurations are provided to give insight into their underlying physics, and the differences in performance were investigated and compared with available literature.

Broughton, Justin (ORCID:000000026848457X)↗
DOE OSTI2021

Correspondence of NNGP Kernel and the Matérn Kernel

Kernels representing limiting cases of neural network architectures have recently gained popularity. However, the application and performance of these new kernels compared to existing options, such as the Matérn kernel, is not well studied. We take a practical approach to explore the neural network Gaussian process (NNGP) kernel and its application to data in Gaussian process regression. We first demonstrate the necessity of normalization to produce valid NNGP kernels and explore related numerical challenges. We further demonstrate that the predictions from this model are quite inflexible, and therefore do not vary much over the valid hyperparameter sets. We then demonstrate a surprising result that the predictions given from the NNGP kernel correspond closely to those given by the Matérn kernel under specific circumstances, which suggests a deep similarity between overparameterized deep neural networks and the Matérn kernel. Finally, we demonstrate the performance of the NNGP kernel as compared to the Matérn kernel on three benchmark data cases, and we conclude that for its flexibility and practical performance, the Matérn kernel is preferred to the novel NNGP in practical applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Clustering and statistical analyses of air-conditioning intensity and use patterns in residential buildings

Energy conservation in residential buildings has gained increased attention due to its large portion of global energy use and potential of energy savings. Occupant behavior has been recognized as a key factor influencing the energy use and load diversity in buildings, therefore more realistic and accurate air-conditioning (AC) operating schedules are imperative for load estimation in equipment design and operation optimization. With the development of sensor technology, it became easier to access an increasing amount of heating/cooling data from thermal energy metering systems in residential buildings, which provides another possible way to understand building energy usage and occupant behaviors. However, except for cooling energy consumption benchmarking, there currently lacks effective and easy approaches to analyze AC usage and provide actionable insights for occupants. To fill this gap, this study proposes clustering analysis to identify AC use patterns of residential buildings, and develops new key performance indicators (KPIs) and data analytics to explore the AC operation characteristics using the long-term metered cooling energy use data, which is of great importance for inhabitants to understand their thermal energy use and save energy cost through adjustment of their AC use behavior. We demonstrate the proposed approaches in a residential district comprising 300 apartments, located in Zhengzhou, China. Main outcomes include: Representative AC use patterns are developed for three room types of residential buildings in the cold climate zone of China, which can be used as more realistic AC schedules to improve accuracy of energy simulation; Distributions of KPIs on household cooling energy usage are established, which can be used for household AC use intensity benchmarking and performance diagnoses.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2023

Evidence of Zintl Intermediate Phase and Its Impacts on Li and Na Storage Performance of Pb-Based Alloying Anodes

Anode materials based on conversion and alloying reactions are promising to achieve high energy density of advanced sodium-ion batteries (SIBs). While the chemical similarities between sodium and lithium as alkali elements make the benchmarking strategy practical in developing new high-performance anodes, simply borrowing the anode material from one system to the other does not always guarantee success unless it is based on sound understanding of both Li- and Na-reaction mechanisms. In this work, we report the Na storage performance of a Pb-based anode and its fundamental reaction dynamics. In contrast to its excellent electrochemical performances in Li cells (reversible ~600 mAh/g), the newly developed Pb@PbO–C nanocomposite anode has limited electrochemical Na reaction properties showing moderate capacity and rate performances (~300 mAh/g at 20 mA/g). Synchrotron-based X-ray diffraction and absorption spectroscopy studies reveal the fundamental differences in the Na and Li reaction mechanism of the Pb-based anode. Further, unlike Li reaction, the unique Na reaction mechanism involves the formation of a highly ionic NaPb Zintl phase, which comprises tetrahedral Pb 4 clusters, as an intermediate phase. The strong covalent character of the Pb 4 Zintl clusters adversely affects the electronic conductivity and thus limits the electrochemical performance of the Pb-based anode in Na cells. These findings provide new insights applicable to developing high-performance alloying anode materials.

25 ENERGY STORAGE↗
NASA NTRS1998

Performance Modeling and Measurement of Parallelized Code for Distributed Shared Memory Multiprocessors

This paper presents a model to evaluate the performance and overhead of parallelizing sequential code using compiler directives for multiprocessing on distributed shared memory (DSM) systems. With increasing popularity of shared address space architectures, it is essential to understand their performance impact on programs that benefit from shared memory multiprocessing. We present a simple model to characterize the performance of programs that are parallelized using compiler directives for shared memory multiprocessing. We parallelized the sequential implementation of NAS benchmarks using native Fortran77 compiler directives for an Origin2000, which is a DSM system based on a cache-coherent Non Uniform Memory Access (ccNUMA) architecture. We report measurement based performance of these parallelized benchmarks from four perspectives: efficacy of parallelization process; scalability; parallelization overhead; and comparison with hand-parallelized and -optimized version of the same benchmarks. Our results indicate that sequential programs can conveniently be parallelized for DSM systems using compiler directives but realizing performance gains as predicted by the performance model depends primarily on minimizing architecture-specific data locality overhead.

Waheed, Abdul↗
NASA NTRS1987

Full CI benchmark calculations on CH3

Full CI calculations have been performed on the CH3 radical. The full CI results are compared to those obtained using CASSCF/multireference CI and coupled-pair functional methods, both at the equilibrium CH distance and at geometries with the three CH bonds extended. In general, the performance of the approximate methods is similar to that observed in calculations on other molecules in which one or two bonds were stretched.

Bauschlicher, Charles W., Jr.↗
DOE OSTI2021

Validation of SAS4A/SASSYS-1 for predicting steady-state single-phase natural circulation

The validation of system analysis codes for nuclear reactor systems is required for the development and application of these computational tools. Designed as a comprehensive system analysis code for advanced nuclear reactors, SAS4A/SASSYS-1 requires validation of its physics model for capturing single-phase natural circulation behavior. To support the validation of SAS4A/SASSYS-1, high-precision experiments are performed capturing steady-state single-phase natural circulation on a scaled facility with comprehensive instrumentation. In this work, dedicated tests are performed quantifying the critical modeling parameters, and a single-phase natural circulation benchmark dataset is obtained with well-documented uncertainty and comprehensive facility description. The validation is then performed against the dataset examining the capability of SAS4A/SASSYS-1 in simulating steady-state single-phase natural circulation. The experimental facility is modeled in the candidate code. Solution verification is performed using Richardson-extrapolation-based estimators which quantify and restrict numerical errors from discretization. Input uncertainty provided by the benchmark dataset is forward propagated through the candidate code, quantifying the output uncertainty in a Monte Carlo approach. The composition of the output uncertainty is also quantified through a variance-based sensitivity analysis. With the uncertainty quantified for each individual condition, a detailed comparison between the simulation results and experimental data is performed covering the whole dataset. The results show consistent agreement for all primary parameters. The current validation activity provides a valuable benchmark dataset for the validation of system analysis codes in capturing single-phase natural circulation and demonstrates satisfactory prediction capability of SAS4A/SASSYS1 for steady-state single-phase natural circulation.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2016

Panda: A Compiler Framework for Concurrent CPU $+$ GPU Execution of 3D Stencil Computations on GPU-accelerated Supercomputers

We present a new compiler framework for truly heterogeneous 3D stencil computation on GPU clusters. Our framework consists of a simple directive-based programming model and a tightly integrated source-to-source compiler. Annotated with a small number of directives, sequential stencil C codes can be automatically parallelized for large-scale GPU clusters. The most distinctive feature of the compiler is its capability to generate hybrid MPI$+$ CUDA$+$ OpenMP code that uses concurrent CPU$+$ GPU computing to unleash the full potential of powerful GPU clusters. The auto-generated hybrid codes hide the overhead of various data motion by overlapping them with computation. Test results on the Titan supercomputer and the Wilkes cluster show that auto-translated codes can achieve about 90 % of the performance of highly optimized handwritten codes, for both a simple stencil benchmark and a real-world application in cardiac modeling. The user-friendliness and performance of our domain-specific compiler framework allow harnessing the full power of GPU-accelerated supercomputing without painstaking coding effort.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

Compiler-based code generation and autotuning for geometric multigrid on GPU-accelerated supercomputers

GPUs, with their high bandwidths and computational capabilities are an increasingly popular target for scientific computing. Unfortunately, to date, harnessing the power of the GPU has required use of a GPU-specific programming model like CUDA, OpenCL, or OpenACC. Thus, in order to deliver portability across CPU-based and GPU-accelerated supercomputers, programmers are forced to write and maintain two versions of their applications or frameworks. In this paper, we explore the use of a compiler-based autotuning framework based on CUDA-CHiLL to deliver not only portability, but also performance portability across CPU- and GPU-accelerated platforms for the geometric multigrid linear solvers found in many scientific applications. We also show that with autotuning we can attain near Roofline (a performance bound for a computation and target architecture) performance across the key operations in the miniGMG benchmark for both CPU- and GPU-based architectures as well as for a multiple stencil discretizations and smoothers. We show that our technology is readily interoperable with MPI resulting in performance at scale equal to that obtained via hand-optimized MPI+CUDA implementation.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

How efficiently can AI recognize Wireless Devices?

This poster presents a hardware benchmarking methodology for a 3-layer CNN waveform classifier deployed using ONNX Runtime on an NVIDIA Jetson AGX Orin. The dataset consist of 9 signal types, -30 to +30 dB SNR with 5dB increments. Benchmarking on the Jetson AGX Orin gave an accuracy of 91.9% and GPU throughput of 107,120 predictions/sec (23× faster than CPU). The Jetson GPU reached approximately 27M samples/sec with stable performance but fell below the 40 MHz rate needed for real-time radio feeds. Sustained testing of 5 minutes confirmed stable performance with no memory leaks, establishing a reproducible benchmarking baseline for future edge-deployment optimization.

99 - GENERAL AND MISCELLANEOUS↗
NASA NTRS1991

Supercomputing '91; Proceedings of the 4th Annual Conference on High Performance Computing, Albuquerque, NM, Nov. 18-22, 1991

Various papers on supercomputing are presented. The general topics addressed include: program analysis/data dependence, memory access, distributed memory code generation, numerical algorithms, supercomputer benchmarks, latency tolerance, parallel programming, applications, processor design, networks, performance tools, mapping and scheduling, characterization affecting performance, parallelism packaging, computing climate change, combinatorial algorithms, hardware and software performance issues, system issues. (No individual items are abstracted in this volume)

Source record↗
DOE OSTI2022

Testing nuclear data libraries with burnup for reactor applications

Recently US and European nuclear data libraries have been released, namely ENDF/B-VIII.0 and JEFF-3.3 libraries, which are the result of years of evaluation and validation work in both communities. Consequently, efforts have been made to validate the evaluations in calculations of integral experiments (critical benchmarks, ..etc). Nevertheless, less stringent testing and validation efforts were performed on burnup applications before releasing the libraries. The presented work focusses on the testing of these recent nuclear data libraries for burnup calculations on two benchmarks at the pin and at the assembly level. Monte-Carlo depletion calculations were performed using the VESTA 2.2 code. The K{sub ∞} results between nuclear data libraries are compared. A strong k{sub ∞} bias is observed with burnup using both JEFF-3.3 and JEFF-4T0 compared to all other libraries, and especially ENDF/B-VIII.0, consisting in a strong k{sub ∞} over-estimation at low burnup and a high under-estimation at high burnup. JEFF-3.3 {sup 235}U and {sup 239}Pu evaluations mainly explain this result, as well as fission yields. New {sup 235}U and {sup 239}Pu evaluations were proposed for JEFF-4T0, but they do not address totally the bias issue, even if JEFF-4T0 {sup 239}Pu allows slightly reducing the k{sub ∞} underestimation at high burnup.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2021

Autotuning PolyBench benchmarks with LLVM Clang/Polly loop optimization pragmas using Bayesian optimization

Here, we develop a ytopt autotuning framework that leverages Bayesian optimization to explore the parameter space search and compare four different supervised learning methods within Bayesian optimization and evaluate their effectiveness. We select six of the most complex PolyBench benchmarks and apply the newly developed LLVM Clang/Polly loop optimization pragmas to the benchmarks to optimize them. We then use the autotuning framework to optimize the pragma parameters to improve their performance. The experimental results show that our autotuning approach outperforms the other compiling methods to provide the smallest execution time for the benchmarks syr2k, 3mm, heat-3d, lu, and covariance with two large datasets in 200 code evaluations for effectively searching the parameter spaces with up to 170,368 different configurations. We find that the Floyd-Warshall benchmark did not benefit from autotuning. To cope with this issue, we provide some compiler option solutions to improve the performance. Then we present loop autotuning without a user's knowledge using a simple mctree autotuning framework to further improve the performance of the Floyd-Warshall benchmark. We also extend the ytopt autotuning framework to tune a deep learning application.

79 ASTRONOMY AND ASTROPHYSICS↗
NASA NTRS2002

Space Operations Training Concepts Benchmark Study (Training in a Continuous Operations Environment)

The NASA/USAF Benchmark Space Operations Training Concepts Study will perform a comparative analysis of the space operations training programs utilized by the United States Air Force Space Command with those utilized by the National Aeronautics and Space Administration. The concentration of the study will be focused on Ground Controller/Flight Controller Training for the International Space Station Payload Program. The duration of the study is expected to be five months with report completion by 30 June 2002. The U.S. Air Force Space Command was chosen as the most likely candidate for this benchmark study because their experience in payload operations controller training and user interfaces compares favorably with the Payload Operations Integration Center's training and user interfaces. These similarities can be seen in the dynamics of missions/payloads, controller on-console requirements, and currency/proficiency challenges to name a few. It is expected that the report will look at the respective programs and investigate goals of each training program, unique training challenges posed by space operations ground controller environments, processes of setting up controller training programs, phases of controller training, methods of controller training, techniques to evaluate adequacy of controller knowledge and the training received, and approaches to training administration. The report will provide recommendations to the respective agencies based on the findings. Attached is a preliminary outline of the study. Following selection of participants and an approval to proceed, initial contact will be made with U.S. Air Force Space Command Directorate of Training to discuss steps to accomplish the study.

Johnston, Alan E.↗