Search NASA⌕ Search

SEARCH · Search NASA

Results for “computer system 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 811 records · Page 45

IER - 551: Experiment for Unresolved Resonance of Plutonium Actinides (EUROPA): CED-1 for True Intermediate Pu Benchmark

IER-551 or the Experiment for Unresolved Resonance Of Plutonium Actinides (EUROPA), is an experiment campaign to design and execute a series of true intermediate energy plutonium critical configurations. EUROPA is a collaboration between Los Alamos National Laboratory (LANL), Oak Ridge National Laboratory (ORNL), and the French Institut de Radioprotection et de Sûreté Nucléaire (IRSN) to validate current and future evaluations of 239,240 Pu isotopes. Preliminary designs were optimized using Particle Swarm Optimization to save on computation time and the configuration with highest percentage of intermediate neutrons causing fission found was BeO-Cd. This combination has over 70% of fissions occurring from intermediate energy neutrons. Secondary considerations regarding end benchmark uncertainties and quality of nuclear data resulted in a beryllium metal only experiment proposed as the final system. This final preliminary design achieves 68% intermediate fissions, sensitivities of fission and capture cross sections of 239 Pu of 0.3021 and -0.1610 respectively.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Potential of minicomputer/array-processor system for nonlinear finite-element analysis

The potential of using a minicomputer/array-processor system for the efficient solution of large-scale, nonlinear, finite-element problems is studied. A Prime 750 is used as the host computer, and a software simulator residing on the Prime is employed to assess the performance of the Floating Point Systems AP-120B array processor. Major hardware characteristics of the system such as virtual memory and parallel and pipeline processing are reviewed, and the interplay between various hardware components is examined. Effective use of the minicomputer/array-processor system for nonlinear analysis requires the following: (1) proper selection of the computational procedure and the capability to vectorize the numerical algorithms; (2) reduction of input-output operations; and (3) overlapping host and array-processor operations. A detailed discussion is given of techniques to accomplish each of these tasks. Two benchmark problems with 1715 and 3230 degrees of freedom, respectively, are selected to measure the anticipated gain in speed obtained by using the proposed algorithms on the array processor.

Strohkorb, G. A.↗

Coolant side heat transfer with rotation. Task 3 report: Application of computational fluid dynamics

An experimental and analytical program was conducted to investigate heat transfer and pressure losses in rotating multipass passages with configurations and dimensions typical of modern turbine blades. The objective of this program is the development and verification of improved analysis methods that will form the basis for a design system that will produce turbine components with improved durability. As part of this overall program, a technique is developed for computational fluid dynamics. The specific objectives were to: select a baseline CFD computer code, assess the limitations of the baseline code, modify the baseline code for rotational effects, verify the modified code against benchmark experiments in the literature, and to identify shortcomings in the code as revealed by the verification. The Pratt and Whitney 3D-TEACH CFD code was selected as the vehicle for this program. The code was modified to account for rotating internal flows, and these modifications were evaluated for flow characteristics of those expected in the application. Results can make a useful contribution to blade internal cooling.

Kopper, F. C.↗

Impact of NASA’s Entry Systems Modeling Project on Planetary Mission Design

Planetary missions continue to grow larger and more complex. Furthermore, the current focus on human exploration of the Moon and Mars, as well as Mars Sample Return(MSR), place increas-ingly stringent requirements on the reliability of the entry, descent, and landing (EDL) system that ensures the safe delivery of payload or crew to their destination. Planetary EDL is an area in which mission designers are critically reliant on modeling and simulation to demonstrate the reliability of the system, as there are no ground facilities that are able to fully test these systems in a flight-relevant environment. NASA’s state-of-the-art modeling and simulation capability must continually evolve to meet the needs of the next generation of planetary EDL. To accomplish this aim, NASA’s Entry Systems Modeling (ESM) Project was formed in 2013and is funded bythe Space Technology Mission Directorate(STMD) and Science Mission Directorate (SMD). ESM is the Agency’s only cross-cutting effort for advancing entry systems modeling and simulation capabilities across a range of technical disciplines and Solar System destinations. ESM is a portfolio project covering a variety of mid-TRL research efforts within four core EDL-related areas of investment: (1) Thermal protection material modeling, (2) Shock layer kinetics and radiation, (3) Aerosciences, and (4) Guidance, navigation, and con-trol. The material modeling group creates detailed material response modelsof thermal protection systems (TPS)from the micro to macro scale, and at the fun-damental and engineering levels. Shock layer kinetics and radiation focuses on radiative heating of space-craft, quantum chemistry and benchmark experiments for validation.Aerosciences is a broad research area that impacts many aspects of entry systems, including parachutes, aerodynamics, and turbulent heating augmentation due to TPS roughness.The guidance, navigation,and control effort under ESM is expanding the capabilities of NASA’s main flight mechanics tool, POST2, for use on high-performance computing architectures and to generalize interoperability with external applications for more detailed end-to-end simula-tion.In addition, several focusedresearch topics have been approvedto augment ESM’s core portfolio. These include efforts for deep post-flight analysis of Mars 2020/MEDLI2 flight data; development ofTPS failure models; improvement of hypersonic wakeflow models; and a recently concluded effort to provide material response models for NuSil-coated PICA heat-shield material. This presentation will discuss each of these investment areas and demonstrate via real mission examples how advances to the state-of-the-art enabled by ESM are directly impacting the missions of today and tomorrow, including InSight, Mars 2020, Mars Sample Return, Orion, and Dragonfly.

Entry Systems Modeling↗

In Situ Synthesis of Uranium Carbide and its High Temperature Cubic Phase

New in situ data for the U-C system are presented, with the goal of improving knowledge of the phase diagram to enable production of new ceramic fuels. The none quenchable, cubic, δ-phase, which in turn is fundamental to computational methods, was identified. Rich datasets of the formation synthesis of uranium carbide yield kinetics data which allow the benchmarking of modeling, thermodynamic parameters etc. The order-disorder transition (carbon sublattice melting) was observed due to equal sensitivity of neutrons to both elements. This dynamic has not been accurately described in some recent simulation-based publications.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

High-Fidelity Measurements for Flattop-HEU Benchmark Reevaluation

Flattop was first built in the 1950’s at Los Alamos National Laboratory. Flattop-HEU is composed of a sphere of highly enriched uranium (HEU) surrounded by a thick spherical natural uranium (NU) reflector. The reflector is composed of three parts: a stationary hemisphere and two movable quarter spheres. For fine control of the reactivity of the system, there are three control rods of natural uranium located in voids in the stationary hemisphere. The final components that make Flattop a useful critical assembly are the glory hole and mass adjustment pieces. These pieces can be loaded in various configurations into the glory hole and the core pedestal to control the known worth of the system. The glory hole and mass adjustment pieces are mostly small pieces of HEU with some mass adjustment pieces fabricated from NU. This allows for the irradiation of samples to a specified level. To better document the system, Flattop was evaluated and included in the International Criticality Safety Benchmark Evaluation Project (ICSBEP) handbook. The original benchmark evaluation of Flattop-HEU was written in 1999 based on an experiment completed in the 1960’s. This original evaluation was written to provide a single diameter that defined critical mass; however, as computational capabilities have increased, the focus for benchmark evaluations has shifted to include detailed modelswith all physical dimensions. Thus, as Flattop is a lynchpin in critical experiment work, the benchmark is being reevaluated at current standards. This summary discusses some of the largest known uncertainties from the evaluation and the high-fidelity measurements taken to reduce these uncertainties.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Development of one-dimensional computational fluid dynamics code 'GFLOW' for groundwater flow and contaminant transport analysis

Prediction of groundwater movement and contaminant transport in soil is an important problem in many branches of science and engineering. This includes groundwater hydrology, environmental engineering, soil science, agricultural engineering and also nuclear engineering. Specifically, in nuclear engineering it is applicable in the design of spent fuel storage pools and waste management sites in the nuclear power plants. Ground water modeling involves the simulation of flow and contaminant transport by groundwater flow. In the context of contaminated soil and groundwater system, numerical simulations are typically used to demonstrate compliance with regulatory standard. A one-dimensional Computational Fluid Dynamics code GFLOW had been developed based on the Finite Difference Method for simulating groundwater flow and contaminant transport through saturated and unsaturated soil. The code is validated with the analytical model and the benchmarking cases available in the literature. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

A benchmark study of electronic excitation energies, transition moments, and excited-state energy gradients on the nicotine molecule

In this work, we report a comparative study of computed excitation energies, oscillator strengths, and excited-state energy gradients of (S)-nicotine, chosen as a test case, using multireference methods, coupled cluster singles and doubles, and methods based on time-dependent density functional theory. This system was chosen because its apparent simplicity hides a complex electronic structure, as several different types of valence excitations are possible, including n-π{sup *}, π-π{sup *}, and charge-transfer states, and in order to simulate its spectrum it is necessary to describe all of them consistently well by the chosen method.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Dynamic Load Balancing on Single- and Multi-GPU Systems

The computational power provided by many-core graphics processing units (GPUs) has been exploited in many applications. The programming techniques supported and employed on these GPUs are not sufficient to address problems exhibiting irregular, and unbalanced workload. The problem is exacerbated when trying to effectively exploit multiple GPUs, which are commonly available in many modern systems. In this paper, we propose a task-based dynamic load-balancing solution for single- and multi-GPU systems. The solution allows load balancing at a finer granularity than what is supported in existing APIs such as NVIDIA’s CUDA. We evaluate our approach using both micro-benchmarks and a molecular dynamics application that exhibits significant load imbalance. Experimental results with a single-GPU configuration show that our fine-grained task solution can utilize the hardware more efficiently than the CUDA scheduler for unbalanced workload. On multi-GPU systems, our solution achieves near-linear speedup, load balance, and significant performance improvement over techniques based on standard CUDA APIs.

97 MATHEMATICS AND COMPUTING↗

Assessment of Zero Power Critical Experiments and Needs for a Fission Surface Power System

The National Aeronautics and Space Administration (NASA) is providing funding to the Department of Energy to assess, develop, and test nuclear technologies that could provide surface power to a lunar outpost. Sufficient testing of this fission surface power (FSP) system will need to be completed to enable a decision by NASA for flight development. The near-term goal for the FSP work is to conduct the minimum amount of testing needed to validate the system performance within an acceptable risk. This paper provides an assessment of the current modeling capabilities and quantifies a preliminary bias associated with the modeling methods for designing the nuclear reactor. Advanced analysis techniques using Zero Power Plutonium Reactor (ZPPR)-20C data should provide sufficient information to preclude the necessity of a cold critical test of the FSP. Further testing to reduce uncertainties in the beryllium and uranium cross-section data should reduce the overall uncertainty in the computational models.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning

Computational modeling of chemical and biological systems at atomic resolution is a crucial tool in the chemist’s toolset. The use of computer simulations requires a balance between cost and accuracy: quantum-mechanical methods provide high accuracy but are computationally expensive and scale poorly to large systems, while classical force fields are cheap and scalable, but lack transferability to new systems. Machine learning can be used to achieve the best of both approaches. Here we train a general-purpose neural network potential (ANI-1ccx) that approaches CCSD(T)/CBS accuracy on benchmarks for reaction thermochemistry, isomerization, and drug-like molecular torsions. This is achieved by training a network to DFT data then using transfer learning techniques to retrain on a dataset of gold standard QM calculations (CCSD(T)/CBS) that optimally spans chemical space. The resulting potential is broadly applicable to materials science, biology, and chemistry, and billions of times faster than CCSD(T)/CBS calculations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Spatio-temporal Fourier Transformer for Long-term Dynamics Prediction (StFT) v1.0

We propose a novel machine learning model spatio-temporal Fourier transformer (StFT) to emulate long-term dynamics of multi-scale and multi-physics systems. Our method StFT overcomes the limitations of rapid error accumulation, particularly in long-term forecasting of systems characterized by complex and coupled dynamics. StFT achieves outstanding accuracy and computational efficiency by effectively capturing multi-scale interactions, and quantify the uncertainties inherent in the predictions. Our model leverages a structured hierarchy of StFT blocks, and explicitly captures dynamics across both macro- and micro- spatial scales. Evaluations conducted on three benchmark datasets (plasma, fluid, and atmospheric dynamics) demonstrate the advantages of our approach over state-of-the-art ML methods.

Bai, Zhe [Lawrence Berkeley National Laboratory (L↗

Extending the POSIX I/O interface: a parallel file system perspective.

The POSIX interface does not lend itself well to enabling good performance for high-end applications. Extensions are needed in the POSIX I/O interface so that high-concurrency HPC applications running on top of parallel file systems perform well. This paper presents the rationale, design, and evaluation of a reference implementation of a subset of the POSIX I/O interfaces on a widely used parallel file system (PVFS) on clusters. Experimental results on a set of micro-benchmarks confirm that the extensions to the POSIX interface greatly improve scalability and performance.

97 MATHEMATICS AND COMPUTING↗

Towards excitations and dynamical quantities in correlated lattices with density matrix embedding theory

Density matrix embedding theory (DMET) provides a framework to describe ground-state expectation values in strongly correlated systems, but its extension to dynamical quantities is still an open problem. We show one route to obtaining excitations and dynamical spectral functions by using the techniques of DMET to approximate the matrix elements that arise in a single-mode inspired excitation ansatz. We demonstrate this approach in the one-dimensional Hubbard model, comparing the neutral excitations, single-particle density of states, charge, and spin dynamical structure factors to benchmarks from the Bethe ansatz and density matrix renormalization group. Finally, our work highlights the potential of these ideas in building computationally efficient approaches for dynamical quantities.

1-dimensional systems↗

SCALE 6.2.4 Validation: Nuclear Criticality Safety

The computational bias of criticality safety computer codes must be established through the validation of the codes to critical experiments. A large collection of suitable experiments has been vetted by the International Criticality Safety Benchmark Evaluation Project (ICSBEP) and made available in the International Handbook of Evaluated Criticality Safety Benchmark Experiments (ICSBEP Handbook). More than 600 cases from this handbook have been prepared and reviewed within the Verified, Archived Library of Inputs and Data (VALID), which is maintained by the Reactor and Nuclear Systems Division at Oak Ridge National Laboratory. The performance of the KENO V.a and KENO-VI Monte Carlo codes within the SCALE 6.2.4 code system is assessed using the VALID models of benchmark experiments. A range of nuclear cross section libraries based on Evaluated Nuclear Data File (ENDF)/B-VII.1 in both multigroup (MG) and continuous energy (CE) formats is considered. The critical experiments available to validate the KENO V.a code cover 15 broad categories of systems. These systems use a range of fissile materials, including a range of uranium enrichments, various plutonium isotopic vectors, and some mixed uranium/plutonium oxides. The physical forms of the fissile material also vary and are represented as metal, solutions, or arrays of rods or plates in a water moderator. The neutron energy spectra of the systems also vary and cover fast, intermediate, mixed, and thermal spectra. Over 550 of the total cases use the KENO V.a code for the four nuclear data libraries considered in this report.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Parallel Agent-Based Simulations on Clusters of GPUs and Multi-Core Processors

An effective latency-hiding mechanism is presented in the parallelization of agent-based model simulations (ABMS) with millions of agents. The mechanism is designed to accommodate the hierarchical organization as well as heterogeneity of current state-of-the-art parallel computing platforms. We use it to explore the computation vs. communication trade-off continuum available with the deep computational and memory hierarchies of extant platforms and present a novel analytical model of the tradeoff. We describe our implementation and report preliminary performance results on two distinct parallel platforms suitable for ABMS: CUDA threads on multiple, networked graphical processing units (GPUs), and pthreads on multi-core processors. Message Passing Interface (MPI) is used for inter-GPU as well as inter-socket communication on a cluster of multiple GPUs and multi-core processors. Results indicate the benefits of our latency-hiding scheme, delivering as much as over 100-fold improvement in runtime for certain benchmark ABMS application scenarios with several million agents. This speed improvement is obtained on our system that is already two to three orders of magnitude faster on one GPU than an equivalent CPU-based execution in a popular simulator in Java. Thus, the overall execution of our current work is over four orders of magnitude faster when executed on multiple GPUs.

97 MATHEMATICS AND COMPUTING↗

Leveraging Prior Concept Learning Improves Generalization From Few Examples in Computational Models of Human Object Recognition

Humans quickly and accurately learn new visual concepts from sparse data, sometimes just a single example. The impressive performance of artificial neural networks which hierarchically pool afferents across scales and positions suggests that the hierarchical organization of the human visual system is critical to its accuracy. These approaches, however, require magnitudes of order more examples than human learners. We used a benchmark deep learning model to show that the hierarchy can also be leveraged to vastly improve the speed of learning. We specifically show how previously learned but broadly tuned conceptual representations can be used to learn visual concepts from as few as two positive examples; reusing visual representations from earlier in the visual hierarchy, as in prior approaches, requires significantly more examples to perform comparably. These results suggest techniques for learning even more efficiently and provide a biologically plausible way to learn new visual concepts from few examples.

Rule, Joshua S.↗

Computational fluid dynamic analysis of a novel particle-to-air fluidized-bed heat exchanger for particle-based thermal energy storage applications

Long-duration energy storage technologies are being targeted to enable cost-effective, decarbonized energy systems. Particle-based thermal energy storage systems are one promising technology by storing excess electricity or heat as sensible thermal energy in inexpensive, solid, inert particles. These systems are only possible if an effective and economical particle-to-working fluid heat exchanger exists. This study predicts the performance of a proposed, direct-contact, particle-to-air, pressurized fluidized-bed heat exchanger using computational fluid dynamics. The common Eulerian-Eulerian framework for modeling fluidized beds is first benchmarked to experimental results at a previously untested operating condition and application. Then, the benchmarked model evaluates the performance of a proposed design for a commercial-scale version of the novel particle-to-air heat exchanger. The results show pressure drop and gas-phase approach temperatures are advantageous compared to other proposed designs for particle-to-air heat exchangers in the literature; approach temperatures were less than 5 °C and gas-phase pressure drop across the fluidized bed was 32 kPa. The model also highlights the importance of gas distributor design and representation in the Eulerian-Eulerian framework to control fluidization behavior. In conclusion, the model built and benchmarked in this study can be leveraged to advance the design and analysis of these heat exchangers critical to the deployment of a promising long-duration energy storage technology.

25 ENERGY STORAGE↗