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 1,747 records · Page 97

Microcanonical Hamiltonian Monte Carlo

We develop Microcanonical Hamiltonian Monte Carlo (MCHMC), a class of models that follow fixed energy Hamiltonian dynamics, in contrast to Hamiltonian Monte Carlo (HMC), which follows canonical distribution with different energy levels. MCHMC tunes the Hamiltonian function such that the marginal of the uniform distribution on the constant-energy-surface over the momentum variables gives the desired target distribution. We show that MCHMC requires occasional energy-conserving billiard-like momentum bounces for ergodicity, analogous to momentum resampling in HMC. We generalize the concept of bounces to a continuous version with partial direction preserving bounces at every step, which gives energy-conserving underdamped Langevin-like dynamics with non-Gaussian noise (MCLMC). MCHMC and MCLMC exhibit favorable scalings with condition number and dimensionality. We develop an efficient hyperparameter tuning scheme that achieves high performance and consistently outperforms NUTS HMC on several standard benchmark problems, in some cases by orders of magnitude.

97 MATHEMATICS AND COMPUTING↗

sKokkos: Enabling Kokkos with Transparent Device Selection on Heterogeneous Systems using OpenACC

This paper presents a new feature to enable Kokkos with transparent device selection. For application developers, it is not easy toidentify which device is the most appropriate to use in a heterogeneous system, since this depends on the characteristics of both the application and the hardware. In Kokkos, a backend is associated with one specific programming model/hardware. Programmers decide which backend to use at compilation time. This new feature implemented on the OpenACC backend eliminates the burden of deciding which device to use, providing a highly productive programming solution for Kokkos applications. This work includes implementation details and a performance study conducted with a set of mini-benchmarks (i.e., AXPY and dot product), kernels (Lattice-Bolzmann method), and two mini-apps (LULESH and miniFE) on two heterogeneous systems with different hardware capabilities. This new Kokkos feature provides high accelerations of up to 35× thanks to automatic and transparent device selection.

Lee, Seyong↗

Development of a ZnS (Ag)-Silicon Photomultiplier Fast Neutron Detector for the Fast Neutron Hodoscope at TREAT

The Transient Reactor Test (TREAT) Facility at Idaho National Laboratory is being restored to support the mission of advanced reactor and fuel development in the United States. A key part of this restoration is the fast neutron hodoscope which images the fuel under test during extreme transient conditions. Traditionally, this has been comprised of ZnS (Ag)-based proton recoil scintillators (PRS) coupled to photomultiplier tubes (PMTs), but silicon photomultipliers (SiPMs) have emerged as a promising candidate to replace PMTs in this application. This work develops, tests, and evaluates a SiPM-based PRS detector for incorporation into the updated Fuel Motion Monitoring System (FMMS) at TREAT. The primary characteristic to be optimized is the gamma rejection efficiency of the detector which will be done using pulse shape discrimination (PSD). To do so multiple types of SiPMs will be tested with multiple PRS buttons and compared using standard PSD figures of merit. Consideration to count rates will also be given and consequently to the detector’s electrical time constants. Once the ideal SiPM has been selected for the PRS detectors, its gamma rejection efficiency will be characterized, and benchmarked to the traditional PMT-PRS detectors. PSD performance for this detector will also be characterized as a function of temperature.

46 - INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AN↗

Cluster Dynamics Modeling Needs for the Advanced Materials and Manufacturing Technologies Program

This milestone report aims to identify and assess the cluster dynamics (CD) modeling requirements within the Department of Energy's Office of Nuclear Energy (DOE-NE) Advanced Materials and Manufacturing Technologies (AMMT) program and to communicate these needs to the DOE-NE Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. The goal is to ensure NEAMS is well-informed about the CD modeling requirements to support AMMT's mission of accelerating the development, qualification, demonstration, and deployment of advanced structural materials and manufacturing for nuclear energy applications. CD modeling is an essential tool for predicting the degradation of structural materials under irradiation, which is a key component of AMMT's accelerated qualification process. The AMMT program focuses on both additively manufactured and wrought structural alloys, such as laser powder-bed fusion 316H austenitic stainless steel, alloy 709, Haynes 244, and alloy 617. These materials require a generalized CD modeling framework to facilitate rapid model development and computational simulation. A flexible, generalized CD software, similar to the Multiphysics Object-Oriented Simulation Environment (MOOSE) finite element framework, would enable modeling of various cluster types, including defect clusters, defect-solute clusters, and multicomponent clusters, incorporating thermodynamics and kinetics parameters. Radiation effects, microstructural feature evolution, and multi-dimensional modeling are critical considerations for the CD model. The usability of the CD code should allow for easy modification and coupling with MOOSE-based simulations. Additionally, the software should adhere to Nuclear Quality Assurance-1 standards, include a testing suite for verification and validation, and be version-controlled within a national laboratory-managed Git repository. Benchmark problems are needed to assess code predictions and performance.

11 - NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Denovo--A New Three-Dimensional Parallel Discrete Ordinates Code in SCALE

Denovo is a new, three-dimensional, discrete ordinates (SN) transport code that uses state-of-the-art solution methods to obtain accurate solutions to the Boltzmann transport equation. Denovo uses the Koch-Baker-Alcouffe parallel sweep algorithm to obtain high parallel efficiency on O(100) processors on XYZ orthogonal meshes. As opposed to traditional SN codes that use source iteration, Denovo uses nonstationary Krylov methods to solve the within-group equations. Krylov methods are far more efficient than stationary schemes. Additionally, classic acceleration schemes (diffusion synthetic acceleration) do not suffer stability problems when used as a preconditioner to a Krylov solver. Denovo's generic programming framework allows multiple spatial discretization schemes and solution methodologies. Denovo currently provides diamond-difference, theta-weighted diamond-difference, linear-discontinuous finite element, trilinear-discontinuous finite element, and step characteristics spatial differencing schemes. Also, users have the option of running traditional source iteration instead of Krylov iteration. Multigroup upscatter problems can be solved using Gauss-Seidel iteration with transport, two-grid acceleration. A parallel first-collision source is also available. Denovo solutions to the Kobayashi benchmarks are in excellent agreement with published results. Parallel performance shows excellent weak scaling up to 20000 cores and good scaling up to 40000 cores.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Walmart Experimental Store Performance Stories: Preprint

Walmart opened two experimental stores--one in Colorado and one in Texas--in 2005 to serve as test beds for several advanced building systems. Each embodied more than 50 experiments covering materials, water systems, energy systems, and renewable energy production. Walmart worked for three years with the National Renewable Energy Laboratory for the Colorado Store and Oak Ridge National Laboratoryfor the Texas store to monitor, analyze, and report on their performance. HVAC experiments included waste oil boilers, a microturbine/absorption chiller combined heat and power system, evaporative cooling, and a transpired solar collector. The refrigeration systems integrated a medium-temperature secondary loop, evaporatively cooled condenser, doors on medium-temperature cases, andlight-emitting diodes on cases. Experiments in the lighting systems included a redesigned roof for clerestory daylighting and T-5 fluorescent lamps. Three photovoltaic systems for a total of 135 kW and a 50-kW wind turbine are also included. The energy system performance was compared to the measured performance of a prototypical Walmart store and to other benchmarks.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Regen: An object layout regenerator on large-scale production HPC systems

This article proposes an object layout regenerator called Regen which regenerates and removes the object layout dynamically to improve the read performance of applications. Regen first detects frequent access patterns from the I/O requests of the applications. Second, Regen reorganizes the objects and regenerates or preallocates new object layouts according to the identified access patterns. Finally, Regen removes or reuses the obsolete or regenerated object layouts as necessary. As a result, Regen accelerates access to objects by providing a flexible object layout. We implement Regen as a framework on top of Proactive Data Container (PDC) and evaluate it on Cori supercomputer, a production-scale HPC system, by using realistic HPC I/O benchmarks. The experimental results show that Regen improves the I/O performance by up to 16.92 × compared with an existing system.

Distributed file system↗

Implementing Molecular Dynamics on Hybrid High Performance Computers - Particle-Particle Particle-Mesh

The use of accelerators such as graphics processing units (GPUs) has become popular in scientific computing applications due to their low cost, impressive floating-point capabilities, high memory bandwidth, and low electrical power requirements. Hybrid high-performance computers, machines with nodes containing more than one type of floating-point processor (e.g. CPU and GPU), are now becoming more prevalent due to these advantages. In this paper, we present a continuation of previous work implementing algorithms for using accelerators into the LAMMPS molecular dynamics software for distributed memory parallel hybrid machines. In our previous work, we focused on acceleration for short-range models with an approach intended to harness the processing power of both the accelerator and (multi-core) CPUs. To augment the existing implementations, we present an efficient implementation of long-range electrostatic force calculation for molecular dynamics. Specifically, we present an implementation of the particle-particle particle-mesh method based on the work by Harvey and De Fabritiis. We present benchmark results on the Keeneland InfiniBand GPU cluster. We provide a performance comparison of the same kernels compiled with both CUDA and OpenCL. We discuss limitations to parallel efficiency and future directions for improving performance on hybrid or heterogeneous computers.

97 MATHEMATICS AND COMPUTING↗

Nuclear Data Testing Of Pu{sup 239} CIELO Evaluation With NDaST

Within the Collaborative International Evaluated Library Organisation (CIELO) project a coordinated effort is being made amongst the different nuclear data evaluation projects, ENDF, JEFF, JENDL etc. to improve the cross sections of six of the most important isotopes, Pu{sup 239}, U{sup 235}, U{sup 238}, Fe{sup 56}, O{sup 16} and H{sup 1}. Recently developed, the Nuclear Data and Sensitivity Tool (NDaST) is an online tool (http://www.oecd-nea.org/ndast/) that allows nuclear data evaluators to get rapid feedback about the consequences of changes to nuclear data on integral experiment benchmarks that exist within the International Criticality Safety Benchmark Evaluation Project (ICSBEP) and International Reactor Physics Evaluation (IRPhE) Project handbooks. NDaST links the thousands of nuclear data sensitivity profiles in the Database for ICSBEP (DICE) and the IRPhE Database and Analysis Tool (IDAT), with the nuclear data viewing tool JANIS. NDaST can be used to provide timely feedback to CIELO participants. Normally an evaluator changes the ENDF/B files based on differential data, and subsequently hands the new evaluation to testers who give feedback on the performance of the new evaluation in terms of accurately predicting the integral benchmark results. This process is often slow. With accurate sensitivity coefficients it is feasible to predict the results of changes to differential data without performing the direct substitution, by simply multiplying the change in cross section with the sensitivity. Furthermore, when multiple aspects of the file are changed simultaneously, the impact of each individual change can be muddled in the substitution testing. NDaST when used in conjunction with direct perturbation testing can help provide feedback as to which cross section changes are driving changes in the benchmarks. In this paper we focus on Pu{sup 239}, and comparing the NDaST predictions of CIELO files to direct substitution. (authors)

07 ISOTOPE AND RADIATION SOURCES↗

Benchmark Simulations of the Thermal-Hydraulic Responses during EBR-II Inherent Safety Tests using SAM

An advanced system analysis tool SAM is being developed for fast-running, improved-fidelity, and whole-plant transient analyses at Argonne National Laboratory under DOE-NE’s Nuclear Energy Advanced Modeling and Simulation (NEAMS) program. As an important part of code development, companion validation activities are being conducted to ensure the performance and validity of the SAM code. This paper presents the benchmark simulations of two EBR-II tests, SHRT-45R and BOP-302R, whose data are available through the support of DOE-NE’s Advanced Reactor Technology (ART) program. The code predictions of major primary coolant system parameter are compared with the test results. Additionally, the SAS4A/SASSYS-1 code simulation results are also included for a code-to-code comparison.

Hu, Rui↗

Benchmarking ICRF Full-wave Solvers for ITER

Abstract Benchmarking of full-wave solvers for ICRF simulations is performed using plasma profiles and equilibria obtained from integrated self-consistent modeling predictions of four ITER plasmas. One is for a high performance baseline (5.3 T, 15 MA) DT H-mode. The others are for half-field, half-current plasmas of interest for the pre-activation phase with bulk plasma ion species being either hydrogen or He4. The predicted profiles are used by six full-wave solver groups to simulate the ICRF electromagnetic fields and heating, and by three of these groups to simulate the current-drive. Approximate agreement is achieved for the predicted heating power for the DT and He4 cases. Factor of two disagreements are found for the cases with second harmonic He3 heating in bulk H cases. Approximate agreement is achieved simulating the ICRF current drive.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

How Does Your Data Center Measure Up? Energy Efficiency Metrics and Benchmarks for Data Center Infrastructure Systems

Data centers are among the most energy intensive types of facilities, and they are growing dramatically in terms of size and intensity [EPA 2007]. As a result, in the last few years there has been increasing interest from stakeholders - ranging from data center managers to policy makers - to improve the energy efficiency of data centers, and there are several industry and government organizations that have developed tools, guidelines, and training programs. There are many opportunities to reduce energy use in data centers and benchmarking studies reveal a wide range of efficiency practices. Data center operators may not be aware of how efficient their facility may be relative to their peers, even for the same levels of service. Benchmarking is an effective way to compare one facility to another, and also to track the performance of a given facility over time. Toward that end, this article presents the key metrics that facility managers can use to assess, track, and manage the efficiency of the infrastructure systems in data centers, and thereby identify potential efficiency actions. Most of the benchmarking data presented in this article are drawn from the data center benchmarking database at Lawrence Berkeley National Laboratory (LBNL). The database was developed from studies commissioned by the California Energy Commission, Pacific Gas and Electric Co., the U.S. Department of Energy and the New York State Energy Research and Development Authority.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Revisiting Temporal Blocking Stencil Optimizations

Iterative stencils are used widely across the spectrum of High Performance Computing (HPC) applications. Many efforts have been put into optimizing stencil GPU kernels, given the prevalence of GPU-accelerated supercomputers. To improve the data locality, temporal blocking is an optimization that combines a batch of time steps to process them together. Under the observation that GPUs are evolving to resemble CPUs in some aspects, we revisit temporal blocking optimizations for GPUs. We explore how temporal blocking schemes can be adapted to the new features in the recent Nvidia GPUs, including large scratchpad memory, hardware prefetching, and device-wide synchronization. We propose a novel temporal blocking method, EBISU, which champions low device occupancy to drive aggressive deep temporal blocking on large tiles that are executed tile-by-tile. We compare EBISU with state-of-the-art temporal blocking libraries: STENCILGEN and AN5D. We also compare with state-of-the-art stencil auto-tuning tools that are equipped with temporal blocking optimizations: ARTEMIS and DRSTENCIL. Over a wide range of stencil benchmarks, EBISU achieves speedups up to 2.53x and a geometric mean speedup of 1.49x over the best state-of-the-art performance in each stencil benchmark.

Zhang, Lingqi↗

Performance Enhancement of APW+lo Calculations by Simplest Separation of Concerns

Full-potential linearized augmented plane wave (LAPW) and APW plus local orbital (APW+lo) codes differ widely in both their user interfaces and in capabilities for calculations and analysis beyond their common central task of all-electron solution of the Kohn–Sham equations. However, that common central task opens a possible route to performance enhancement, namely to offload the basic LAPW/APW+lo algorithms to a library optimized purely for that purpose. To explore that opportunity, we have interfaced the Exciting-Plus (“EP”) LAPW/APW+lo DFT code with the highly optimized SIRIUS multi-functional DFT package. This simplest realization of the separation of concerns approach yields substantial performance over the base EP code via additional task parallelism without significant change in the EP source code or user interface. We provide benchmarks of the interfaced code against the original EP using small bulk systems, and demonstrate performance on a spin-crossover molecule and magnetic molecule that are of size and complexity at the margins of the capability of the EP code itself.

Zhang, Long↗

Performance Analysis of Scientific Computing Workloads on Trusted Execution Environments

Scientific computing sometimes involves computation on sensitive data. Depending on the data and the execution environment, the HPC (high-performance computing) user or data provider may require confidentiality and/or integrity guarantees. To study the applicability of hardware-based trusted execution environments (TEEs) to enable secure scientific computing, we deeply analyze the performance impact of AMD SEV and Intel SGX for diverse HPC benchmarks including traditional scientific computing, machine learning, graph analytics, and emerging scientific computing workloads. We observe three main findings: 1) SEV requires careful memory placement on large scale NUMA machines (1x -3.4x slowdown without and 1x -1.15x slowdown with NUMA aware placement), 2) virtualization - a prerequisite for SEV - results in performance degradation for workloads with irregular memory accesses and large working sets (1x -4x slowdown compared to native execution for graph applications) and 3) SGX is inappropriate for HPC given its limited secure memory size and inflexible programming model (1.2x -126x slowdown over unsecure execution). Finally, we discuss forthcoming new TEE designs and their potential impact on scientific computing.

97 MATHEMATICS AND COMPUTING↗

BENCHMARK EVALUATION OF THE START-UP CORE REACTOR PHYSICS MEASUREMENTS OF THE HIGH TEMPERATURE ENGINEERING TEST REACTOR

The benchmark evaluation of the start-up core reactor physics measurements performed with Japan’s High Temperature Engineering Test Reactor, in support of the Next Generation Nuclear Plant Project and Very High Temperature Reactor Program activities at the Idaho National Laboratory, has been completed. The evaluation was performed using MCNP5 with ENDF/B-VII.0 nuclear data libraries and according to guidelines provided for inclusion in the International Reactor Physics Experiment Evaluation Project Handbook. Results provided include updated evaluation of the initial six critical core configurations (five annular and one fully-loaded). The calculated keff eigenvalues agree within 1s of the benchmark values. Reactor physics measurements that were evaluated include reactivity effects measurements such as excess reactivity during the core loading process and shutdown margins for the fully-loaded core, four isothermal temperature reactivity coefficient measurements for the fully-loaded core, and axial reaction rate measurements in the instrumentation columns of three core configurations. The calculated values agree well with the benchmark experiment measurements. Fully subcritical and warm critical configurations of the fully-loaded core were also assessed. The calculated keff eigenvalues for these two configurations also agree within 1s of the benchmark values. The reactor physics measurement data can be used in the validation and design development of future High Temperature Gas-cooled Reactor systems.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Establishing the quantum supremacy frontier with a 281 Pflop/s simulation

Noisy intermediate-scale quantum (NISQ) computers are entering an era in which they can perform computational tasks beyond the capabilities of the most powerful classical computers, thereby achieving 'quantum supremacy', a major milestone in quantum computing. NISQ supremacy requires comparison with a state-of-the-art classical simulator. In this paper, we report HPC simulations of hard random quantum circuits (RQC), which have been recently used as a benchmark for the first experimental demonstration of quantum supremacy, sustaining an average performance of 281 Pflop/s (true single precision) on Summit, currently the fastest supercomputer in the world. These simulations were carried out using qFlex, a tensor-network-based classical high-performance simulator of RQCs. Our results show an advantage of many orders of magnitude in energy consumption of NISQ devices over classical supercomputers. In addition, we propose a standard benchmark for NISQ computers based on qFlex.

97 MATHEMATICS AND COMPUTING↗

Application of the EPRI depletion reactivity benchmarks - 14423

The EPRI measured-based depletion reactivity benchmarks intend to facilitate the validation of burnup credit by validating the combined effect of changes in isotopic content and cross-sections. In the framework of burnup credit implementation for PWR fuels, this paper proposes to study the application of the EPRI benchmarks for a spent fuel pool criticality evaluation using the French tools: the DARWIN 2.2 depletion package and the CRISTAL V1 criticality safety package, both based on cross-section libraries from the JEF2.2 evaluation. For that purpose, the benchmarks are analyzed using the connection of DARWIN 2.2 with APOLLO2-MORET 4, allowing the determination of the bias for the reactivity decrement. Then, these results are used to define a bias to take into account in the criticality evaluation of a spent fuel pool. In addition, the benefit of the EPRI depletion reactivity benchmarks is assessed through a comparison with the validation approach adopted in France until now. The bias resulting from this latter approach is evaluated by 1) applying, to the calculated nuclides concentrations, isotopic penalizing factors derived from the comparison between calculated values of nuclide concentrations and experimental values obtained from used fuel chemical analyses from French reactors, 2) analyzing representative critical experiments. Additional analysis of the benchmarks using the connection of VESTA 2.1 with MORET 5 is performed as these codes allow calculations of complex 3D geometrical configurations and the use of various nuclear data libraries. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗