Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory benchmark”

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 217 records · Page 12

Use of SCALE Continuous-Energy Monte Carlo Tools for Eigenvalue Sensitivity Coefficient Calculations

The TSUNAMI code within the SCALE code system makes use of eigenvalue sensitivity coefficients for an extensive number of criticality safety applications, such as quantifying the data-induced uncertainty in the eigenvalue of critical systems, assessing the neutronic similarity between different critical systems, and guiding nuclear data adjustment studies. The need to model geometrically complex systems with improved fidelity and the desire to extend TSUNAMI analysis to advanced applications has motivated the development of a methodology for calculating sensitivity coefficients in continuous-energy (CE) Monte Carlo applications. The CLUTCH and Iterated Fission Probability (IFP) eigenvalue sensitivity methods were recently implemented in the CE KENO framework to generate the capability for TSUNAMI-3D to perform eigenvalue sensitivity calculations in continuous-energy applications. This work explores the improvements in accuracy that can be gained in eigenvalue and eigenvalue sensitivity calculations through the use of the SCALE CE KENO and CE TSUNAMI continuous-energy Monte Carlo tools as compared to multigroup tools. The CE KENO and CE TSUNAMI tools were used to analyze two difficult models of critical benchmarks, and produced eigenvalue and eigenvalue sensitivity coefficient results that showed a marked improvement in accuracy. The CLUTCH sensitivity method in particular excelled in terms of efficiency and computational memory requirements.

97 MATHEMATICS AND COMPUTING↗

Baseline Assessment of TREAT for Modeling and Analysis Needs

TREAT is an air-cooled, graphite moderated, thermal, heterogeneous test facility designed to evaluate reactor fuels and structural materials under conditions simulating various types of nuclear excursions and transient undercooling situations that could occur in a nuclear reactor. After 21 years in a standby mode, TREAT is being re-activated to revive transient testing capabilities. Given the time elapsed and the concurrent loss of operating experience, current generation and advanced computational methods are being applied to begin TREAT modeling and simulation prior to renewed at-power operations. Such methods have limited value in predicting the behavior of TREAT without proper validation. Hence, the U.S. DOE has developed a number of programs to support development of benchmarks for both critical and transient operations. Extensive effort has been expended at INL to collect detailed descriptions, drawings and specifications for all aspects of TREAT, and to resolve conflicting data found through this process. This report provides a collection of these data, with updated figures that are significantly more readable than historic drawings and illustrations, compositions, and dimensions based on the best available sources. This document is not nor should it be considered to be a benchmark report. Rather, it is intended to provide one-stop shopping, to the extent possible, for other work that seeks to prepare detailed, accurate models of the core and its components. Given the nature of the variety of historic documents available and the loss of institutional memory, the only completely accurate database of TREAT data is TREAT itself. Unfortunately, disassembly of TREAT for inspection, assay, and measurement is highly unlikely. Hence the data provided herein is intended serve as a best-estimate substitute.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Roofline model toolkit: A practical tool for architectural and program analysis

We present preliminary results of the Roofline Toolkit for multicore, many core, and accelerated architectures. This paper focuses on the processor architecture characterization engine, a collection of portable instrumented micro benchmarks implemented with Message Passing Interface (MPI), and OpenMP used to express thread-level parallelism. These benchmarks are specialized to quantify the behavior of different architectural features. Compared to previous work on performance characterization, these microbenchmarks focus on capturing the performance of each level of the memory hierarchy, along with thread-level parallelism, instruction-level parallelism and explicit SIMD parallelism, measured in the context of the compilers and run-time environments. We also measure sustained PCIe throughput with four GPU memory managed mechanisms. By combining results from the architecture characterization with the Roofline model based solely on architectural specifications, this work offers insights for performance prediction of current and future architectures and their software systems. To that end, we instrument three applications and plot their resultant performance on the corresponding Roofline model when run on a Blue Gene/Q architecture.

97 MATHEMATICS AND COMPUTING↗

A new paradigm for local-global coupling in whole-core neutron transport.

A new paradigm that increases the efficiency of whole-core neutron transport calculations without lattice homogenization is introduced. Quasi-reflected interface conditions are formulated to partially decouple periodic lattice effects from global flux gradients. The starting point is the finite subelement form of the variational nodal code VARIANT that eliminates fuel-coolant homogenization through the use of heterogeneous nodes. The interface spherical harmonics expansions that couple pin-cell-sized nodes are divided into low-order and high-order terms, and reflected interface conditions are applied to the high-order terms. Combined with an integral transport method within the node, the new approach dramatically reduces both the formation time and the dimensions of the nodal response matrices and leads to sharply reduced memory requirements and computational time. The method is applied to the two-dimensional C5G7 problem, an Organisation for Economic Co-operation and Development/Nuclear Energy Agency pressurized water reactor benchmark containing mixed oxide (MOX) and UO{sub 2} fuel assemblies, as well as to a three-dimensional MOX fuel assembly. Results indicate the new approach results in very little loss of accuracy relative to the corresponding full spherical harmonics expansions while reducing computational times by well over an order of magnitude.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

FPGA-based HPC accelerators: An evaluation on performance and energy efficiency

Hardware specialization is a promising direction for the future of digital computing. Reconfigurable technologies enable hardware specialization with modest non-recurring engineering cost, but their performance and energy efficiency compared to state-of-the-art processor architectures remain an open question. In this article, we use FPGAs to evaluate the benefits of building specialized hardware for numerical kernels found in scientific applications. In order to properly evaluate performance, we not only compare Intel Arria 10 and Xilinx U280 performance against Intel Xeon, Intel Xeon Phi, and NVIDIA V100 GPUs, but we also extend the Empirical Roofline Toolkit (ERT) to FPGAs in order to assess our results in terms of the Roofline model. We show design optimization and tuning techniques for peak FPGA performance at reasonable hardware usage and power consumption. As FPGA peak performance is known to be far less than that of a GPU, we also benchmark the energy efficiency of each platform for the scientific kernels comparing against microbenchmark and technological limits. Results show that while FPGAs struggle to compete in absolute terms with GPUs on memory- and compute-intensive kernels, they require far less power and can deliver nearly the same energy efficiency.

97 MATHEMATICS AND COMPUTING↗

s-Step Krylov Subspace Methods as Bottom Solvers for Geometric Multigrid

Geometric multigrid solvers within adaptive mesh refinement (AMR) applications often reach a point where further coarsening of the grid becomes impractical as individual sub domain sizes approach unity. At this point the most common solution is to use a bottom solver, such as BiCGStab, to reduce the residual by a fixed factor at the coarsest level. Each iteration of BiCGStab requires multiple global reductions (MPI collectives). As the number of BiCGStab iterations required for convergence grows with problem size, and the time for each collective operation increases with machine scale, bottom solves in large-scale applications can constitute a significant fraction of the overall multigrid solve time. In this paper, we implement, evaluate, and optimize a communication-avoiding s-step formulation of BiCGStab (CABiCGStab for short) as a high-performance, distributed-memory bottom solver for geometric multigrid solvers. This is the first time s-step Krylov subspace methods have been leveraged to improve multigrid bottom solver performance. We use a synthetic benchmark for detailed analysis and integrate the best implementation into BoxLib in order to evaluate the benefit of a s-step Krylov subspace method on the multigrid solves found in the applications LMC and Nyx on up to 32,768 cores on the Cray XE6 at NERSC. Overall, we see bottom solver improvements of up to 4.2x on synthetic problems and up to 2.7x in real applications. This results in as much as a 1.5x improvement in solver performance in real applications.

97 MATHEMATICS AND COMPUTING↗

Detection and Diagnosis of Data Integrity Attacks in Solar Farms Based on Multilayer Long Short-Term Memory Network

Photovoltaic (PV) systems are becoming more vulnerable to cyber threats. In response to this emerging concern, developing cyber-secure power electronics converters has received increased attention from the IEEE Power Electronics Society that recently launched a cyber-physical-security initiative. Here this letter proposes a deep sequence learning based diagnosis solution for data integrity attacks on PV systems in smart grids, including dc–dc and dc–ac converters. The multilayer long short-term memory networks are used to leverage time-series electric waveform data from current and voltage sensors in PV systems. The proposed method has been evaluated in a PV smart grid benchmark model with extensive quantitative analysis. As a comparison, we have evaluated classic data-driven methods, including K-nearest neighbor, decision tree, support vector machine, artificial neural network, and convolutional neural network. Comparison results verify performances of the proposed method for detection and diagnosis of various data integrity attacks on PV systems.

42 ENGINEERING↗

Using SciDB to Support Photon Science Data Analysis

Array data analytic systems like SciDB hold great potential to accelerate processing data from SLAC's Linac Coherent Light Source and other experiments. SciDB is unique in its ability to integrate storage and processing of array data efficiently, providing both space-efficient storage and out-of-memory efficient parallel array processing. We describe a recent effort to leverage SciDB to store and process LCLS data. The work includes development of software to import data into SciDB, subsequent benchmarks, and interactive manipulation of data in SciDB.

43 PARTICLE ACCELERATORS↗

Final Report

The purpose of this project was to develop tools and techniques to improve the ability of computational scientists to investigate and correct problems (bugs) in their programs. Specifically, the University of Maryland component of this project focused on the problems associated with the finite number of bits available in a computer to represent numeric values. In large scale scientific computation, numbers are frequently added to and multiplied with each other billions of times. Thus even small errors due to the representation of numbers can accumulate into big errors. However, using too many bits to represent a number results in additional computation, memory, and energy costs. Thus it is critical to find the right size for numbers. This project focused on several aspects of this general problem. First, we developed a tool to look for cancelations, the catastrophic loss of precision in numbers due to the addition of two numbers whose actual values are close to each other, but whose representation in a computer is identical or nearly so. Second, we developed a suite of tools to allow programmers to identify exactly how much precision is required for each operation in their program. This tool allows programmers to both verify that enough precision is available, but more importantly find cases where extra precision could be eliminated to allow the program to use less memory, computer time, or energy. These tools use advanced binary modification techniques to allow the analysis of actual optimized code. The system, called Craft, has been applied to a number of benchmarks and real applications.

97 MATHEMATICS AND COMPUTING↗

Generalized master equation for particle transport in binary random media with renewal statistics

Particle transport in binary stochastic mixtures is classically modeled assuming Markovian or exponential mixing statistics but in many applications material memory invalidates the Markov assumption. For non-Markovian mixing characterized by alternating renewal processes, a transport-theoretic framework is presented that provides an exact description of transport in nonscattering random binary media with general non-exponential statistics. Our approach is to Markovianize the problem by augmenting the {material type, particle flux} state space with the age or distance from the last interface. A Chapman-Kolmogorov equation is formulated for the joint probability density of the material type, particle flux, and age, and subsequently reduced to a generalized Master equation (GME) in differential form. This constitutes the primary result of this work. A state-updating Monte Carlo algorithm consistent with the GME is developed and benchmarked against analytical solutions for multiple chord-length laws. For purely absorbing renewal statistical media, the GME reproduces analytical benchmarks for the equilibrium age distribution, interior mean/variance of material-conditioned fluxes, and boundary transmittance. Simulations further demonstrate that a Markov (exponential) approximation of non-exponential statistics can introduce large errors in transmittance and interior flux profiles. Lastly, the reintroduction of memory due to scattering is briefly addressed through heuristic considerations.

Fluctuations & noise↗

Tools for Simulation and Benchmark Generation at Exascale

The path to exascale high-performance computing (HPC) poses several challenges related to power, performance, resilience, productivity, programmability, data movement, and data management. Investigating the performance of parallel applications at scale on future architectures and the performance impact of different architecture choices is an important component of HPC hardware/software co-design. Simulations using models of future HPC systems and communication traces from applications running on existing HPC systems can offer an insight into the performance of future architectures. This work targets technology developed for scalable application tracing of communication events and memory profiles, but can be extended to other areas, such as I/O, control flow, and data flow. It further focuses on extreme-scale simulation of millions of Message Passing Interface (MPI) ranks using a lightweight parallel discrete event simulation (PDES) toolkit for performance evaluation. Instead of simply replaying a trace within a simulation, the approach is to generate a benchmark from it and to run this benchmark within a simulation using models to reflect the performance characteristics of future-generation HPC systems. This provides a number of benefits, such as eliminating the data intensive trace replay and enabling simulations at different scales. The presented work utilizes the ScalaTrace tool to generate scalable trace files, the ScalaBenchGen tool to generate the benchmark, and the xSim tool to run the benchmark within a simulation.

Lagadapati, Mahesh↗

Online thermal profile prediction for large format additive manufacturing: A hybrid CNN-LSTM based approach

Large format additive manufacturing (LFAM) is an advanced 3D printing technique that efficiently fabricates large-scale components through a layer-by-layer extrusion and deposition process. Accurate surface layer temperature monitoring is essential to prevent manufacturing failures and ensure final product quality. Traditional physics-based offline approaches for simulating thermal behavior are often inefficient and complex, posing challenges on real-time, in-situ monitoring. Here, to address this, we propose a data-driven hybrid CNN-LSTM model to predict sequential thermal images of arbitrary length using real-time infrared thermal imaging. In this approach, a Convolutional Neural Networks (CNN) is trained offline to capture spatial features, reduce dimensional complexity, and enhance time efficiency, while a stacked Long Short-Term Memory (LSTM) is applied online to capture temporal information for improved prediction of future thermal behavior in subsequent printing layers. Model performance is evaluated using MSE, SSIM, and PSNR metrics and is benchmarked against stacked LSTM and convolutional LSTM models, demonstrating superior accuracy and applicability. Additionally, to mitigate noise from moving extruders and gantry backgrounds in thermal images, a fine-tuned semantic segmentation model is implemented offline to extract printing geometry, enabling precise temperature tracking along the tool path for further thermal analysis. The frameworks developed in this study significantly advance temperature monitoring, thermal analysis, and in-situ manufacturing control for LFAM, bridging the gap between theoretical modeling and practical application.

Geometry extraction↗

Comb

Comb is a communication performance benchmarking tool. It is used to determine performance tradeoffs in implementing communication patterns on high performance computing (HPC) platforms. At its core comb runs combinations of communication patterns with execution patterns, and memory spaces. The current set of capabilities Comb provides includes: o Configurable structured mesh halo exchange communication. o A variety of communication patterns based on grouping messages. o A variety of execution patterns including serial, openmp threading, cuda, cuda batched kernels, and cuda persistent kernels. o A variety of memory spaces including default system allocated memory, pinned host memory, cuda device memory, and cuda managed memory with different cuda memory advice.

Burmark, Jason↗

Strategies for Energy-Efficient Resource Management of Hybrid Programming Models

Many scientific applications are programmed using hybrid programming models that use both message passing and shared memory, due to the increasing prevalence of large-scale systems with multicore, multisocket nodes. Previous work has shown that energy efficiency can be improved using software-controlled execution schemes that consider both the programming model and the power-aware execution capabilities of the system. However, such approaches have focused on identifying optimal resource utilization for one programming model, either shared memory or message passing, in isolation. The potential solution space, thus the challenge, increases substantially when optimizing hybrid models since the possible resource configurations increase exponentially. Nonetheless, with the accelerating adoption of hybrid programming models, we increasingly need improved energy efficiency in hybrid parallel applications on large-scale systems. In this work, we present new software-controlled execution schemes that consider the effects of dynamic concurrency throttling (DCT) and dynamic voltage and frequency scaling (DVFS) in the context of hybrid programming models. Specifically, we present predictive models and novel algorithms based on statistical analysis that anticipate application power and time requirements under different concurrency and frequency configurations. Furthermore, we apply our models and methods to the NPB MZ benchmarks and selected applications from the ASC Sequoia codes. Overall, we achieve substantial energy savings (8.74 percent on average and up to 13.8 percent) with some performance gain (up to 7.5 percent) or negligible performance loss.

97 MATHEMATICS AND COMPUTING↗

Progress on Optimizing Wind Farms and Rotor Designs Using Adjoints

Modern wind plants are increasingly tasked with multiple performance objectives. In addition to designing plants that maximize power output and minimize the levelized cost of energy (LCOE), the design and operation of wind plants is increasingly influenced by challenges regarding grid integration of variable generation renewables. This places a growing emphasis on making wind plants more controllable and predictable. WindSE is a Reynolds-averaged Navier-Stokes (RANS) model designed around analytical gradient and adjoint methods, with the ability to capture terrain-induced effects, as shown in Figure 1. The recent addition of an unsteady solver with an actuator line method (ALM) and ongoing work to enable massively parallel optimizations gives it a unique niche to explore coupled plant-level controls and design problems. This code is an open source python package built on the FEniCS framework that utilizes fast, parallel PETSc solvers to model fluid flow throughout wind-farm scale domains. Two recent studies performed using WindSE demonstrate the capability to optimize under a wide variety of flow conditions and objective functions. In the first, we present an optimization focused on modifying the layout of a wind farm with a fixed number of turbines for maximum total power output [1]. This study highlights the ability to quickly perform simulations using the steady Navier-Stokes solver combined with rotors represented as actuator disks while also stressing the importance of capturing terrain-induced effects. Gradient-based optimization using the RANS equations is viable due to the inclusion of efficiently computed adjoint derivatives. We interpret the physical results of the optimal layout and also discuss the computational cost of scaling to larger problems. In the second study, we present the capabilities of the unsteady Navier-Stokes solver, where rotor-blade profiles represented by actuator lines are optimized to enhance wake steering effects and overall power production [2]. We quantify the wind plant performance gains obtained from this type of simultaneous control co-design optimization as compared to optimizing the blade design and yaw independently. Figure 2 shows the differences between a baseline two-turbine system and an optimized system where we fine-tune the blade chord profile. Results and challenges from each study are quickly summarized and used to motivate the current development efforts within WindSE. Current and future work is focused on enabling higher-resolution studies with more degrees of freedom through parallelization of both the simulation and optimization algorithms. We present benchmarking results to show that WindSE performs well in both weak- and strong-scaling tests and further demonstrate that the optimizer obtains the same convergence rates in both shared- and distributed-memory environments. Using larger wind farms, we can study deep-array effects within an optimization context, allowing the use of objective functions that have been previously unstudied. As an example, we present ongoing work on a blockage metric which characterizes the loss of available kinetic energy due to wake effects from multiple upstream turbines.

adjoint optimization↗

Verification of High-Fidelity Neutronics Code PROTEUS for C5G7 Benchmark Problems

Under the DOE Nuclear Energy Advanced Modeling and Simulation (NEAMS) program, the 3D whole core transport code PROTEUS has been developed at Argonne National Laboratory to perform the detailed 3D transport calculation for various reactor types including irregular or complex geometric reactors. PROTEUS includes two distinct transport solvers: SN and MOC. Among them, the MOC solver based on the extruded geometry, thus named MOCEX, can produce the faithful 3-D transport solutions by employing the rigorous and robust 2D/3D MOC formulation. This formulation overcomes the drawbacks of a full 3D MOC method requiring enormous memory and computational resources and resolves the instability issues of the 2D/1D method as well. The solution of MOCEX converges to that of a Monte Carlo code as the discretization of angle, space and energy is refined, unlike the 2D/1D method. As a verification and validation (V and V) effort of PROTEUS-MOCEX, this paper presents verification test results using the well-known 3D C5G7 benchmark problems by solving and comparing eigenvalues and pin powers with MCNP solutions.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Tuning collective communication for Partitioned Global Address Space programming models

Partitioned Global Address Space (PGAS) languages offer programmers the convenience of a shared memory programming style combined with locality control necessary to run on large-scale distributed memory systems. Even within a PGAS language programmers often need to perform global communication operations such as broadcasts or reductions, which are best performed as collective operations in which a group of threads work together to perform the operation. In this study we consider the problem of implementing collective communication within PGAS languages and explore some of the design trade-offs in both the interface and implementation. In particular, PGAS collectives have semantic issues that are different than in send–receive style message passing programs, and different implementation approaches that take advantage of the one-sided communication style in these languages. We present an implementation framework for PGAS collectives as part of the GASNet communication layer, which supports shared memory, distributed memory and hybrids. The framework supports a broad set of algorithms for each collective, over which the implementation may be automatically tuned. In conclusion, we demonstrate the benefit of optimized GASNet collectives using application benchmarks written in UPC, and demonstrate that the GASNet collectives can deliver scalable performance on a variety of state-of-the-art parallel machines including a Cray XT4, an IBM BlueGene/P, and a Sun Constellation system with InfiniBand interconnect.

97 MATHEMATICS AND COMPUTING↗

Higher order treatment on temporal derivative of angular flux for time-dependent MOC

A new kinetic analysis method, whose angular dependence of temporal derivative for angular flux is accurately treated within practical memory requirement, is proposed. The method of characteristics (MOC) is being widely used for reactor analysis thanks to the advances of numerical algorithms and computer hardware. However, the computational resources, i.e., the memory capacity, can be still a crucial problem for rigorous kinetic calculations using MOC. In the straightforward approach for kinetic calculation using MOC, the segment-averaged angular fluxes should be stored on the memory in order to explicitly calculate the temporal derivative of the angular flux, which would require huge memory. Thus, in the conventional kinetic calculation code using MOC, the temporal derivative of the angular flux has been approximated as angularly isotropic in order to reduce the memory requirement (isotropic assumption). However, the approximation error caused by the conventional isotropic assumption has not been thoroughly and quantitatively investigated so far and an accurate kinetic calculation method, which can quantitatively estimate the above approximation error within practical memory storage, has not been developed. The present study tries to address this issue with a newly developed approach. Effect of the approximate treatment for the temporal derivative of angular flux is evaluated through benchmark calculations. (authors)

97 MATHEMATICS AND COMPUTING↗