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 1,459 records · Page 81

Self-consistent continuum solvation for optical absorption of complex molecular systems in solution

We introduce a new method to compute the optical absorption spectra of complex molecular systems in solution, based on the Liouville approach to time-dependent density-functional perturbation theory and the revised self-consistent continuum solvation model. The former allows one to obtain the absorption spectrum over a whole wide frequency range, using a recently proposed Lanczos-based technique, or selected excitation energies, using the Casida equation, without having to ever compute any unoccupied molecular orbitals. The latter is conceptually similar to the polarizable continuum model and offers the further advantages of allowing an easy computation of atomic forces via the Hellmann-Feynman theorem and a ready implementation in periodic-boundary conditions. The new method has been implemented using pseudopotentials and plane-wave basis sets, benchmarked against polarizable continuum model calculations on 4-aminophthalimide, alizarin, and cyanin and made available through the QUANTUM ESPRESSO distribution of open-source codes.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Let Each Quantum Bit Choose Its Basis Gates

Near-term quantum computers are primarily limited by errors in quantum operations (or gates) between two quantum bits (or qubits). A physical machine typically provides a set of basis gates that include primitive 2-qubit (2Q) and 1-qubit (1Q) gates that can be implemented in a given technology. 2Q entangling gates, coupled with some 1Q gates, allow for universal quantum computation. In superconducting technologies, the current state of the art is to implement the same 2Q gate between every pair of qubits (typically an XX-or XY-type gate). This strict hardware uniformity requirement for 2Q gates in a large quantum computer has made scaling up a time and resource-intensive endeavor in the lab. We propose a radical idea – allow the 2Q basis gate(s) to differ between every pair of qubits, selecting the best entangling gates that can be calibrated between given pairs of qubits. This work aims to give quantum scientists the ability to run meaningful algorithms with qubit systems that are not perfectly uniform. Scientists will also be able to use a much broader variety of novel 2Q gates for quantum computing. We develop a theoretical framework for identifying good 2Q basis gates on “nonstandard” Cartan trajectories that deviate from “standard” trajectories like XX. We then introduce practical methods for calibration and compilation with nonstandard 2Q gates, and discuss possible ways to improve the compilation. To demonstrate our methods in a case study, we simulated both standard XY-type trajectories and faster, nonstandard trajectories using an entangling gate architecture with far-detuned transmon qubits. We identify efficient 2Q basis gates on these nonstandard trajectories and use them to compile a number of standard benchmark circuits such as QFT and QAOA. Furthermore, our results demonstrate an 8x improvement over the baseline 2Q gates with respect to speed and coherence-limited gate fidelity.

quantum computing↗

Modeling of Cloud Microphysics: Can We Do Better?

Representation of cloud microphysics is a key aspect of simulating clouds. From the early days of cloud modeling, numerical models have relied on an Eulerian approach for all cloud and thermodynamic and microphysics variables. Over time the sophistication of microphysics schemes has steadily increased, from simple representations of bulk masses of cloud and rain in each grid cell, to including different ice particle types and bulk hydrometeor concentrations, to complex schemes referred to as bin or spectral schemes that explicitly evolve the hydrometeor size distributions within each model grid cell. As computational resources grow, there is a clear trend toward wider use of bin schemes, including their use as benchmarks to develop and test simplified bulk schemes. We argue that continuing on this path brings fundamental challenges difficult to overcome. The Lagrangian particle-based probabilistic approach is a practical alternative in which the myriad of cloud and precipitation particles present in a natural cloud is represented by a judiciously selected ensemble of point particles called superdroplets or superparticles. Here, the advantages of the Lagrangian particle-based approach when compared to the Eulerian bin methodology are explained, and the prospects of applying the method to more comprehensive cloud simulations—for instance, targeting deep convection or frontal cloud systems—are discussed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

TChem-atm (v2.0.0): scalable performance-portable multiphase atmospheric chemistry

We present TChem-atm, a performance-portable approach that enables efficient simulation of chemically detailed and multiphase atmospheric chemistry on modern heterogeneous computing architectures. Unlike previous efforts that rely on architecture-specific code or focus exclusively on gas-phase chemistry, TChem-atm supports fully coupled gas–aerosol systems with execution across CPUs, NVIDIA GPUs, and AMD GPUs through the Kokkos programming model. It integrates the flexible multiphase capabilities of the Community Atmospheric Model Chemistry Package (CAMP) with the high-performance kinetic routines of TChem, and includes automatic Jacobian construction with support for a range of stiff ODE solvers. In a proof-of-concept integration with the particle-resolved model PartMC, TChem-atm reproduces the existing PartMC–CAMP implementation within solver tolerances and delivers substantial GPU speedups, especially for large particle populations. Performance benchmarks reveal substantial speedups on GPU platforms, particularly for large particle populations, with consistent results across hardware backends. TChem-atm enables performance-portable execution across CPUs and GPUs, though optimal efficiency may require modest architecture-specific tuning (e.g., team and vector sizes), with up to a twofold improvement on the NVIDIA H100. It directly supports sectional and particle-resolved host models, while modal aerosol schemes require minor adaptation to provide particle-scale quantities such as representative diameters. By enabling chemically detailed, multiphase simulations with performance portability and host-model flexibility, TChem-atm facilitates the incorporation of advanced chemistry into atmospheric models.

Díaz-Ibarra, Oscar Homero [Sandia National Laborat↗

Physical properties of the benchmark models program supercritical wing

The goal of the Benchmark Models Program is to provide data useful in the development and evaluation of aeroelastic computational fluid dynamics (CFD) codes. To that end, a series of three similar wing models are being flutter tested in the Langley Transonic Dynamics Tunnel. These models are designed to simultaneously acquire model response data and unsteady surface pressure data during wing flutter conditions. The supercritical wing is the second model of this series. It is a rigid semispan model with a rectangular planform and a NASA SC(2)-0414 supercritical airfoil shape. The supercritical wing model was flutter tested on a flexible mount, called the Pitch and Plunge Apparatus, that provides a well-defined, two-degree-of-freedom dynamic system. The supercritical wing model and associated flutter test apparatus is described and experimentally determined wind-off structural dynamic characteristics of the combined rigid model and flexible mount system are included.

Dansberry, Bryan E.↗

Effectively Using Remote I/O For Work Composition in Distributed Workflows

Distributed scientific workflows are becoming more important with the interest in incorporating AI into their loops. A critical programming and performance question is how to compose workflow tasks when data is produced on one system but must be consumed on another. Since the dominant technique is composition with remote I/O, this paper explores its performance expectations. We describe BigFlowSim, a workflow I/O simulator that captures key implementation choices for remote I/O, including intensity, reuse, locality, access pattern, and data movement.With BigFlowSim, we generate a synthetic benchmark. We quantify the effects of each parameter with a performance sensitivity study. We explain trends in terms of data movement reduction and show that, under certain conditions, it is possible to establish a total order among most parameters. We apply these insights to a high energy physics workflow, Belle II Monte Carlo and simulate several I/O optimizations. Speedups range from 5% to 2×, without changing compute time.

Friese, Ryan D.↗

Evaluation of the concrete shield compositions from the 2010 criticality accident alarm system benchmark experiments at the CEA Valduc SILENE facility

In October 2010, a series of benchmark experiments were conducted at the French Commissariat a l'Energie Atomique et aux Energies Alternatives (CEA) Valduc SILENE facility. These experiments were a joint effort between the United States Department of Energy Nuclear Criticality Safety Program and the CEA. The purpose of these experiments was to create three benchmarks for the verification and validation of radiation transport codes and evaluated nuclear data used in the analysis of criticality accident alarm systems. This series of experiments consisted of three single-pulsed experiments with the SILENE reactor. For the first experiment, the reactor was bare (unshielded), whereas in the second and third experiments, it was shielded by lead and polyethylene, respectively. The polyethylene shield of the third experiment had a cadmium liner on its internal and external surfaces, which vertically was located near the fuel region of SILENE. During each experiment, several neutron activation foils and thermoluminescent dosimeters (TLDs) were placed around the reactor. Nearly half of the foils and TLDs had additional high-density magnetite concrete, high-density barite concrete, standard concrete, and/or BoroBond shields. CEA Saclay provided all the concrete, and the US Y-12 National Security Complex provided the BoroBond. Measurement data from the experiments were published at the 2011 International Conference on Nuclear Criticality (ICNC 2011) and the 2013 Nuclear Criticality Safety Division (NCSD 2013) topical meeting. Preliminary computational results for the first experiment were presented in the ICNC 2011 paper, which showed poor agreement between the computational results and the measured values of the foils shielded by concrete. Recently the hydrogen content, boron content, and density of these concrete shields were further investigated within the constraints of the previously available data. New computational results for the first experiment are now available that show much better agreement with the measured values.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Scalar Breit interaction for molecular calculations

Variational treatment of the Dirac–Coulomb–Gaunt or Dirac–Coulomb–Breit two-electron interaction at the Dirac–Hartree–Fock level is the starting point of high-accuracy four-component calculations of atomic and molecular systems. In this work, we introduce, for the first time, the scalar Hamiltonians derived from the Dirac–Coulomb–Gaunt and Dirac–Coulomb–Breit operators based on spin separation in the Pauli quaternion basis. While the widely used spin-free Dirac–Coulomb Hamiltonian includes only the direct Coulomb and exchange terms that resemble nonrelativistic two-electron interactions, the scalar Gaunt operator adds a scalar spin–spin term. The spin separation of the gauge operator gives rise to an additional scalar orbit-orbit interaction in the scalar Breit Hamiltonian. Benchmark calculations of Au n (n = 2–8) show that the scalar Dirac–Coulomb–Breit Hamiltonian can capture 99.99% of the total energy with only 10% of the computational cost when real-valued arithmetic is used, compared to the full Dirac–Coulomb–Breit Hamiltonian. In conclusion, the scalar relativistic formulation developed in this work lays the theoretical foundation for the development of high-accuracy, low-cost correlated variational relativistic many-body theory.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Bridging the gap between parallel file systems and local file systems : a case study with PVFS.

Parallel I/O plays an increasingly important role in today's data intensive computing applications. While much attention has been paid to parallel read performance, most of this work has focused on the parallel file system, middleware, or application layers, ignoring the potential for improvement through more effective use of local storage. In this paper, we present the design and implementation of segment-structured on-disk data grouping and prefetching (SOGP), a technique that leverages additional local storage to boost the local data read performance for parallel file systems, especially for those applications with partially overlapped access patterns. Parallel virtual file system (PVFS) is chosen as an example. Our experiments show that an SOGP-enhanced PVFS prototype system can outperform a traditional Linux-Ext3-based PVFS for many applications and benchmarks, in some tests by as much as 230% in terms of I/O bandwidth.

99 GENERAL AND MISCELLANEOUS↗

Co-design of Advanced Architectures for Graph Analytics using Machine Learning

A graph is an excellent way of representing relationships among entities. We can use graph analytics to synthesize and analyze such relational data, and extract relevant features that are useful for various tasks such as machine learning. Considering the crucial role of graph analytics in various domains, it is important and timely to investigate the right hardware configurations that can achieve optimal performance for graph workloads on future high-performance computing systems. Design space exploration studies facilitate the selection of appropriate configurations (e.g. memory) to achieve a desired system performance. Recently, the approach of accelerating graph analytics using persistent non-volatile memory has gained a lot of attention. Traditional system simulators such as Gem5 and NVMain can be used to explore the design space of these advanced memory architectures for graph workloads. However, these simulators are slow in execution thus limiting the efficiency of design space exploration studies. To overcome this challenge, we proposed a machine learning based approach to co-design advanced memory architectures for graph workloads. We tested our approach with DRAM, non-volatile memory, and hybrid memory (DRAM+NVM) using a breadth first search benchmark algorithm. Our results showed the applicability of the proposed machine learning based approach to the co-design of the advanced memory architectures. In this paper, we provide recommendations on selecting advanced memory architectures to achieve desired performance for graph workloads. We also discuss the performances of different machine learning models that were considered in this study.

Kurte, Kuldeep↗

Palacios and Kitten : high performance operating systems for scalable virtualized and native supercomputing.

Palacios and Kitten are new open source tools that enable applications, whether ported or not, to achieve scalable high performance on large machines. They provide a thin layer over the hardware to support both full-featured virtualized environments and native code bases. Kitten is an OS under development at Sandia that implements a lightweight kernel architecture to provide predictable behavior and increased flexibility on large machines, while also providing Linux binary compatibility. Palacios is a VMM that is under development at Northwestern University and the University of New Mexico. Palacios, which can be embedded into Kitten and other OSes, supports existing, unmodified applications and operating systems by using virtualization that leverages hardware technologies. We describe the design and implementation of both Kitten and Palacios. Our benchmarks show that they provide near native, scalable performance. Palacios and Kitten provide an incremental path to using supercomputer resources that is not performance-compromised.

99 GENERAL AND MISCELLANEOUS↗

RELAP5 Model of the First Wall/Blanket Primary Heat Transfer System

ITER inductive power operation is modeled and simulated using a system level computer code to evaluate the behavior of the Primary Heat Transfer System (PHTS) and predict parameter operational ranges. The control algorithm strategy and derivation are summarized in this report as well. A major feature of ITER is pulsed operation. The plasma does not burn continuously, but the power is pulsed with large periods of zero power between pulses. This feature requires active temperature control to maintain a constant blanket inlet temperature and requires accommodation of coolant thermal expansion during the pulse. In view of the transient nature of the power (plasma) operation state a transient system thermal-hydraulics code was selected: RELAP5. The code has a well-documented history for nuclear reactor transient analyses, it has been benchmarked against numerous experiments, and a large user database of commonly accepted modeling practices exists. The process of heat deposition and transfer in the blanket modules is multi-dimensional and cannot be accurately captured by a one-dimensional code such as RELAP5. To resolve this, a separate CFD calculation of blanket thermal power evolution was performed using the 3-D SC/Tetra thermofluid code. A 1D-3D co-simulation more realistically models FW/blanket internal time-dependent thermal inertia while eliminating uncertainties in the time constant assumed in a 1-D system code. Blanket water outlet temperature and heat release histories for any given ITER pulse operation scenario are calculated. These results provide the basis for developing time dependent power forcing functions which are used as input in the RELAP5 calculations.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

GASNet-EX Memory Kinds: Support for Device Memory in PGAS Programming Models

There is an emerging need for adaptive, lightweight communication in irregular HPC applications at exascale, where GPU accelerators provide the majority of available compute cycles. To address this need, Lawrence Berkeley National Lab is developing a programming system to support distributed-memory HPC application development using the Partitioned Global Address Space (PGAS) model. This work includes two major components: UPC++ and GASNet-EX. UPC++ is a C++ template library providing Remote Memory Access (RMA) and Remote Procedure Call (RPC) communication interfaces. GASNet-EX is a portable, high-performance communication middleware library, used by the implementations of UPC++ and many other PGAS programming models. We describe recent advances in GASNet-EX to efficiently implement zero-copy Remote Memory Access (RMA) communication to and from memory on accelerator devices such as GPUs. We demonstrate performance improvements via benchmark results from UPC++ (on Summit) and the Legion programming system (on DGX-1), both using GASNet-EX for communication.

Hargrove, Paul H↗

Ice Accretions and Full-Scale Iced Aerodynamic Performance Data for a Two-Dimensional NACA 23012 Airfoil

This report documents the data collected during the large wind tunnel campaigns conducted as part of the SUNSET project (StUdies oN Scaling EffecTs due to ice) also known as the Ice-Accretion Aerodynamics Simulation study: a joint effort by NASA, the Office National d'Etudes et Recherches Aérospatiales (ONERA), and the University of Illinois. These data form a benchmark database of full-scale ice accretions and corresponding ice-contaminated aerodynamic performance data for a two-dimensional (2D) NACA 23012 airfoil. The wider research effort also included an analysis of ice-contaminated aerodynamics that categorized ice accretions by aerodynamic effects and an investigation of subscale, low- Reynolds-number ice-contaminated aerodynamics for the NACA 23012 airfoil. The low-Reynolds-number investigation included an analysis of the geometric fidelity needed to reliably assess aerodynamic effects of airfoil icing using artificial ice shapes. Included herein are records of the ice accreted during campaigns in NASA Glenn Research Center's Icing Research Tunnel (IRT). Two different 2D NACA 23012 airfoil models were used during these campaigns; an 18-in. (45.7-cm) chord (subscale) model and a 72-in. (182.9-cm) chord (full-scale) model. The aircraft icing conditions used during these campaigns were selected from the Federal Aviation Administration's (FAA's) Code of Federal Regulations (CFR) Part 25 Appendix C icing envelopes. The records include the test conditions, photographs of the ice accreted, tracings of the ice, and ice depth measurements. Model coordinates and pressure tap locations are also presented. Also included herein are the data recorded during a wind tunnel campaign conducted in the F1 Subsonic Pressurized Wind Tunnel of ONERA. The F1 tunnel is a pressured, high- Reynolds-number facility that could accommodate the full-scale (72-in. (182.9-cm) chord) 2D NACA 23012 model. Molds were made of the ice accreted during selected test runs of the full-scale model in the IRT. From these molds, castings were made that closely replicated the features of the accreted ice. The castings were then mounted on the full-scale model in the F1 tunnel, and aerodynamic performance measurements were made using model surface pressure taps, the facility force balance system, and a large wake rake designed specifically for these tests. Tests were run over a range of Reynolds and Mach numbers. For each run, the model was rotated over a range of angles-of-attack that included airfoil stall. The benchmark data collected during these campaigns were, and continue to be, used for various purposes. The full-scale data form a unique, ice-accretion and associated aerodynamic performance dataset that can be used as a reference when addressing concerns regarding the use of subscale ice-accretion data to assess full-scale icing effects. Further, the data may be used in the development or enhancement of both ice-accretion prediction codes and computational fluid dynamic codes when applied to study the effects of icing. Finally, as was done in the wider study, the data may be used to help determine the level of geometric fidelity needed for artificial ice used to assess aerodynamic degradation due to aircraft icing. The structured, multifaceted approach used in this research effort provides a unique perspective on the aerodynamic effects of aircraft icing. The data presented in this report are available in electronic form upon formal approval by proper NASA and ONERA authorities.

Addy, Harold E., Jr.↗

Consistent Second Moment Methods with Scalable Linear Solvers for Radiation Transport

Second moment methods (SMMs) are developed that are consistent with the discontinuous Galerkin spatial discretization of the discrete ordinates (or S\(_N\)) transport equations. The low-order (LO) diffusion system of equations is discretized with fully consistent P\(_1\), local discontinuous Galerkin (LDG), and interior penalty (IP) methods. A discrete residual approach is used to derive SMM correction terms that make each of the LO systems consistent with the high-order discretization. We show that the consistent methods are more accurate and have better solution quality than independently discretized LO systems, that they preserve the diffusion limit, and that the LDG and IP consistent SMMs can be scalably solved in parallel on a challenging, multimaterial benchmark problem.

97 MATHEMATICS AND COMPUTING↗

Advance Resource Provisioning in Bulk Data Scheduling

Today?s scientific and business applications generate mas- sive data sets that need to be transferred to remote sites for sharing, processing, and long term storage. Because of increasing data volumes and enhancement in current net- work technology that provide on-demand high-speed data access between collaborating institutions, data handling and scheduling problems have reached a new scale. In this paper, we present a new data scheduling model with ad- vance resource provisioning, in which data movement operations are defined with earliest start and latest comple- tion times. We analyze time-dependent resource assign- ment problem, and propose a new methodology to improve the current systems by allowing researchers and higher-level meta-schedulers to use data-placement as-a-service, so they can plan ahead and submit transfer requests in advance. In general, scheduling with time and resource conflicts is NP-hard. We introduce an efficient algorithm to organize multiple requests on the fly, while satisfying users? time and resource constraints. We successfully tested our algorithm in a simple benchmark simulator that we have developed, and demonstrated its performance with initial test results.

97 MATHEMATICS AND COMPUTING↗

Analytical harmonic vibrational frequencies with VV10-containing density functionals: Theory, efficient implementation, and benchmark assessments

VV10 is a powerful nonlocal density functional for long-range correlation that is used to include dispersion effects in many modern density functionals, such as the meta-generalized gradient approximation (mGGA), B97M-V, the hybrid GGA, ωB97X-V, and the hybrid mGGA, ωB97M-V. While energies and analytical gradients for VV10 are already widely available, this study reports the first derivation and efficient implementation of the analytical second derivatives of the VV10 energy. The additional compute cost of the VV10 contributions to analytical frequencies is shown to be small in all but the smallest basis sets for recommended grid sizes. Here, this study also reports the assessment of VV10-containing functionals for predicting harmonic frequencies using the analytical second derivative code. The contribution of VV10 to simulating harmonic frequencies is shown to be small for small molecules but important for systems where weak interactions are important, such as water clusters. In the latter cases, B97M-V, ωB97M-V, and ωB97X-V perform very well. The convergence of frequencies with respect to the grid size and atomic orbital basis set size is studied, and recommendations are reported. Finally, scaling factors to allow comparison of scaled harmonic frequencies with experimental fundamental frequencies and to predict zero-point vibrational energy are presented for some recently developed functionals (including r2SCAN, B97M-V, ωB97X-V, M06-SX, and ωB97M-V).

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Performance and energy impact of OpenMP runtime configurations on power constrained systems

Energy consumption of an application has attained the same vitality as execution time in current HPC systems, primarily due to the massive energy consumption of these systems and consequently substantial energy bills. This fact forced software developers as well as hardware vendors to put an extensive focus on energy consumption. This focus becomes even more crucial in power constrained systems, where the amount of power dedicated to applications is limited. In such systems, finding the optimal balance between execution time and energy consumption has a significant impact on utilizing these systems efficiently. A programming model is a necessary layer where optimizations can be applied to find the optimal balance. OpenMP is a standard programming model for intra-node parallelism. In this paper, we give a detailed analysis of how to optimize the execution time and energy consumption of an OpenMP application using the OpenMP runtime environment at different power levels. We studied twenty-eight OpenMP parallel regions from five NAS Parallel Benchmark (NPB) applications. Based on the study, we show that for a given power level, a suitable selection of OpenMP runtime parameters can improve the execution time and energy consumption of a parallel region up to 67% and 72%, respectively. We also show that these fine-grained improvements resulted in up to 26% of execution time and 38% of energy consumption improvement for a given OpenMP application. We finally observed that optimizing the energy consumption behavior alone across all power levels regardless of execution time behavior can lead to an additional energy consumption improvement of up to 55%. However, this improvement comes at the expense of execution time.

97 MATHEMATICS AND COMPUTING↗