Search NASA⌕ Search

SEARCH · Search NASA

Results for “computer system performance”

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 541 records · Page 30
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

ESnet-JLab FPGA Accelerated Transport (control plane) [EJFAT (udplbd2)] v2.0

The ESnet-JLab FPGA Accelerated Transport system is a solution for streaming high-speed scientific measurement data from Data Acquisition Systems (DAQs) to high-performance computing facilties. It is generally compatible with many science workflows, and makes no assumptions about the specifics of any particular experiment. This program (udplbd version 2) implements the control plane for the system. It is responsible for programming network forwarding rules into the data plane (implemented by the hardware designed named udplb, described separately). It also implements the control loop necessary to match up offered workload with available capacity on high-performance compute nodes.

Howard, Derek [Lawrence Berkeley National Laborato↗
DOE OSTI2022

How the Hydrophobic Interface between a Perfluorosulfonic Acid Polymer and Water Vapor Controls Membrane Hydration

Stable hydration in perfluorinated polyelectrolyte membranes such as Nafion is essential to maintain good ion conductivity and manage permeation, especially in vapor-fed devices where water content depends on relative humidity in a gas stream. Extensive studies in the literature have shown that Nafion hydration in water vapor is controlled by its interfacial transport resistance. Nafion forms a fluorine-rich layer at the polymer-gas interface, and it has been proposed that this layer blocks water transport due to its hydrophobicity. To develop a molecular-level description of the physics underlying transport resistance in this system, we have performed a computational reaction-diffusion kinetics study of water evaporation from Nafion. Two distinct models are examined, one mimicking the blocking function proposed in the literature and the other assuming that there is no blocking, treating instead water evaporation as a dynamic balance between uptake from the gas and desorption from the polymer surface. Simulation results are compared to time-dependent infrared data over a range of 100-0% relative humidity from the literature. Only the dynamic model successfully reproduces experimental observations. This indicates that the physical nature of interfacial transport resistance is not slow diffusion across an interfacial layer; rather, it is due to the competition between dehydration and rehydration. The simulation data provide details on the accompanying water distributions throughout the membrane and on interfacial kinetics, showing that they are characterized by strong fluctuations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

Baryon magnetic moments: Symmetries and relations

Magnetic moments of the octet baryons are computed using lattice QCD in background magnetic fields, including the first treatment of the magnetically coupled Σ0- Λ system. Although the computations are performed for relatively large values of the up and down quark masses, we gain new insight into the symmetries and relations between magnetic moments by working at a three-flavor mass-symmetric point. While the spinflavor symmetry in the large Nc limit of QCD is shared by the naïve constituent quark model, we find instances where quark model predictions are considerably favored over those emerging in the large Nc limit. We suggest further calculations that would shed light on the curious patterns of baryon magnetic moments.

Parreno, Assumpta↗
DOE OSTI2015

An atomic orbital-based formulation of the complete active space self-consistent field method on graphical processing units

Despite its importance, state-of-the-art algorithms for performing complete active space self-consistent field (CASSCF) computations have lagged far behind those for single reference methods. We develop an algorithm for the CASSCF orbital optimization that uses sparsity in the atomic orbital (AO) basis set to increase the applicability of CASSCF. Our implementation of this algorithm uses graphical processing units (GPUs) and has allowed us to perform CASSCF computations on molecular systems containing more than one thousand atoms. Additionally, we have implemented analytic gradients of the CASSCF energy; the gradients also benefit from GPU acceleration as well as sparsity in the AO basis.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2016

A singly charged ion source for radioactive {sup 11}C ion acceleration

A new singly charged ion source using electron impact ionization has been developed to realize an isotope separation on-line system for simultaneous positron emission tomography imaging and heavy-ion cancer therapy using radioactive {sup 11}C ion beams. Low-energy electron beams are used in the electron impact ion source to produce singly charged ions. Ionization efficiency was calculated in order to decide the geometric parameters of the ion source and to determine the required electron emission current for obtaining high ionization efficiency. Based on these considerations, the singly charged ion source was designed and fabricated. In testing, the fabricated ion source was found to have favorable performance as a singly charged ion source.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2015

Cryogenic molecular separation system for radioactive {sup 11}C ion acceleration

A {sup 11}C molecular production/separation system (CMPS) has been developed as part of an isotope separation on line system for simultaneous positron emission tomography imaging and heavy-ion cancer therapy using radioactive {sup 11}C ion beams. In the ISOL system, {sup 11}CH{sub 4} molecules will be produced by proton irradiation and separated from residual air impurities and impurities produced during the irradiation. The CMPS includes two cryogenic traps to separate specific molecules selectively from impurities by using vapor pressure differences among the molecular species. To investigate the fundamental performance of the CMPS, we performed separation experiments with non-radioactive {sup 12}CH{sub 4} gases, which can simulate the chemical characteristics of {sup 11}CH{sub 4} gases. We investigated the separation of CH{sub 4} molecules from impurities, which will be present as residual gases and are expected to be difficult to separate because the vapor pressure of air molecules is close to that of CH{sub 4}. We determined the collection/separation efficiencies of the CMPS for various amounts of air impurities and found desirable operating conditions for the CMPS to be used as a molecular separation device in our ISOL system.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2022

Fixed-target serial crystallography at the Structural Biology Center

Serial synchrotron crystallography enables the study of protein structures under physiological temperature and reduced radiation damage by collection of data from thousands of crystals. The Structural Biology Center at Sector 19 of the Advanced Photon Source has implemented a fixed-target approach with a new 3D-printed mesh-holder optimized for sample handling. The holder immobilizes a crystal suspension or droplet emulsion on a nylon mesh, trapping and sealing a near-monolayer of crystals in its mother liquor between two thin Mylar films. Data can be rapidly collected in scan mode and analyzed in near real-time using piezoelectric linear stages assembled in an XYZ arrangement, controlled with a graphical user interface and analyzed using a high-performance computing pipeline. Here, the system was applied to two β-lactamases: a class D serine β-lactamase from Chitinophaga pinensis DSM 2588 and L1 metallo-β-lactamase from Stenotrophomonas maltophilia K279a.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Interstate-24 MOTION: Closing the Loop on Smart Mobility

We introduce the I-24 Mobility Technology Interstate Observation Network (MOTION), a transportation cyber-physical systems testbed under development in Tennessee. It consists of a six-mile freeway segment instrumented with 400 4K resolution cameras, processed by a real-time compute system to enable continuous performance monitoring of freeway traffic. The testbed is being developed to support next generation connected and autonomous vehicle technologies and advanced traffic management. When complete, the testbed will be the longest continuously observed freeway segment in the world. This article introduces the testbed, discusses the core design choices, and outlines the preliminary work conducted to support the design.

Gloudemans, Derek↗
DOE OSTI2017

Integrating External Resources with a Task-Based Programming Model

Accessing external resources (e.g., loading input data, checkpointing snapshots, and out-of-core processing) can have a significant impact on the performance of supercomputer applications. However, no existing programming systems for high-performance computing directly manage and optimize these external accesses. As a result, users must explicitly manage external accesses alongside their computation at the application level, which can result in both correctness and performance issues. We address this limitation by introducing Iris, a task-based programming model with semantics for external resources. Iris allows applications to describe their access requirements to external resources and the relationship of those accesses to the computation. Iris incorporates external I/O into a deferred execution model, reschedules external I/O to overlap I/O with computation, and reduces external I/O when possible. We evaluate Iris on three microbenchmarks representative of important workloads in HPC and a full combustion simulation, S3D. We demonstrate that the Iris implementation of S3D reduces the external I/O overhead by up to 20×, compared to the Legion and the Fortran implementations.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2014

MO-G-17A-01: Innovative High-Performance PET Imaging System for Preclinical Imaging and Translational Researches

Purpose: To develop a practical and compact preclinical PET with innovative technologies for substantially improved imaging performance required for the advanced imaging applications. Methods: Several key components of detector, readout electronics and data acquisition have been developed and evaluated for achieving leapfrogged imaging performance over a prototype animal PET we had developed. The new detector module consists of an 8×8 array of 1.5×1.5×30 mm{sup 3} LYSO scintillators with each end coupled to a latest 4×4 array of 3×3 mm{sup 2} Silicon Photomultipliers (with ∼0.2 mm insensitive gap between pixels) through a 2.0 mm thick transparent light spreader. Scintillator surface and reflector/coupling were designed and fabricated to reserve air-gap to achieve higher depth-of-interaction (DOI) resolution and other detector performance. Front-end readout electronics with upgraded 16-ch ASIC was newly developed and tested, so as the compact and high density FPGA based data acquisition and transfer system targeting 10M/s coincidence counting rate with low power consumption. The new detector module performance of energy, timing and DOI resolutions with the data acquisition system were evaluated. Initial Na-22 point source image was acquired with 2 rotating detectors to assess the system imaging capability. Results: No insensitive gaps at the detector edge and thus it is capable for tiling to a large-scale detector panel. All 64 crystals inside the detector were clearly separated from a flood-source image. Measured energy, timing, and DOI resolutions are around 17%, 2.7 ns and 1.96 mm (mean value). Point source image is acquired successfully without detector/electronics calibration and data correction. Conclusion: Newly developed advanced detector and readout electronics will be enable achieving targeted scalable and compact PET system in stationary configuration with >15% sensitivity, ∼1.3 mm uniform imaging resolution, and fast acquisition counting rate capability for substantially improved imaging and quantification performance for small animal imaging and image-guided radiotherapy applications. This work was supported by a research award RP120326 from Cancer Prevention and Research Institute of Texas.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2014

Poster — Thur Eve — 02: Measurement of CT radiation profile width using Fuji CR imaging plate raw data

Measuring the CT collimation width and assessing the shape of the overall profile is a relatively straightforward quality control (QC) measure that impacts both image quality and patient dose, and is often required at acceptance and routine testing. Most CT facilities have access to computed radiography (CR) systems, so performing CT collimation profile assessments using CR plates requires no additional equipment. Previous studies have shown how to effectively use CR plates to measure the radiation profile width. However, a major limitation of the previous work is that the full dynamic range of CR detector plates are not used, since the CR processing technology reduces the dynamic range of the DICOM output to 2{sup 10}, requiring the sensitivity and latitude settings of CR reader to be adjusted to prevent clipping of the CT profile data. Such adjustments to CR readers unnecessarily complicate the QC procedure. These clipping artefacts hinder the ability to accurately assess CT collimation width because the full-width at half maximum value of the penumbras are not properly determined if the maximum dose of the profile is not available. Furthermore, any inconsistencies in the radiation profile shape are lost if the profile plateau is clipped off. In this work we developed an opensource Matlab script for straightforward CT profile width measurements using raw CR data that also allows assessment of the profile shape without clipping, and applied this approach during CT QC.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2023

Computational Requirements in Clean Energy and Manufacturing: Summary report of the virtual workshop held on June 28-29, 2021

On June 28–29, 2021, the US Department of Energy’s (DOE’s) Advanced Scientific Computing Research (ASCR) program in the Office of Science convened a workshop with the Energy Efficiency and Renew able Energy (EERE) program offices to assess the future need for advanced computing resources in the areas of clean energy and advanced manufacturing. In part, this discussion served as an update to earlier workshops and townhalls. ASCR is guided by DOE mission needs as it develops research programs, computers, and networks at the leading edge of technologies. As the exascale computing era dawns, technology changes are creating new opportunities for those who must use high-performance computing (HPC) and data systems effectively. The ASCR computing facilities are augmenting their strategy to adapt to changing science needs and emerging technologies and to leverage the utility of exascale computing across the federal government.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Extending SST vanadis to Add SIMT Functional Units

Sandia National Laboratories is currently investigating scalable architectural simulation capabilities, with a focus on simulating and evaluating highly scalable supercomputers for high-performance computing applications. This exploration is driven by the shift toward more specialized forms of compute and the need for a more diverse set of accurate models. This project will explore the use of General-Purpose Graphical Processing Units (GPGPUs) in high-performance computing using both physical systems and new simulator models – traditional GPUs as well as tightly-coupled SIMT accelerators.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Parallel Programming in MCNP6

Monte Carlo N-Particle (MCNP)1 is a general-purpose Monte Carlo particle transport code developed by Los Alamos National Laboratory (LANL). To efficiently handle long simulations, MCNP version 6 (MCNP6) supports parallel execution using two primary programming models: • Shared-memory task-based threading using OpenMP (Open Multi-Processing), and • Distributed-memory calculations using MPI (Message Passing Interface). The OpenMP and MPI programming models enable MCNP6 to scale from desktop systems to high-performance computing (HPC) clusters, allowing users to run MCNP in one of three parallel modes: • OpenMP-only, • MPI-only, and • Hybrid (MPI + OpenMP). The choice of parallelization mode depends on the underlying computer architecture and the characteristics of the simulation problem.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Opportunities for enhancing MLCommons efforts while leveraging insights from educational MLCommons earthquake benchmarks efforts

MLCommons is an effort to develop and improve the artificial intelligence (AI) ecosystem through benchmarks, public data sets, and research. It consists of members from start-ups, leading companies, academics, and non-profits from around the world. The goal is to make machine learning better for everyone. In order to increase participation by others, educational institutions provide valuable opportunities for engagement. In this article, we identify numerous insights obtained from different viewpoints as part of efforts to utilize high-performance computing (HPC) big data systems in existing education while developing and conducting science benchmarks for earthquake prediction. As this activity was conducted across multiple educational efforts, we project if and how it is possible to make such efforts available on a wider scale. This includes the integration of sophisticated benchmarks into courses and research activities at universities, exposing the students and researchers to topics that are otherwise typically not sufficiently covered in current course curricula as we witnessed from our practical experience across multiple organizations. As such, we have outlined the many lessons we learned throughout these efforts, culminating in the need for benchmark carpentry for scientists using advanced computational resources. The article also presents the analysis of an earthquake prediction code benchmark while focusing on the accuracy of the results and not only on the runtime; notedly, this benchmark was created as a result of our lessons learned. Energy traces were produced throughout these benchmarks, which are vital to analyzing the power expenditure within HPC environments. Additionally, one of the insights is that in the short time of the project with limited student availability, the activity was only possible by utilizing a benchmark runtime pipeline while developing and using software to generate jobs from the permutation of hyperparameters automatically. It integrates a templated job management framework for executing tasks and experiments based on hyperparameters while leveraging hybrid compute resources available at different institutions. The software is part of a collection called cloudmesh with its newly developed components, cloudmesh-ee (experiment executor) and cloudmesh-cc (compute coordinator).

58 GEOSCIENCES↗
DOE OSTI2014

List based prefetch

A list prefetch engine improves a performance of a parallel computing system. The list prefetch engine receives a current cache miss address. The list prefetch engine evaluates whether the current cache miss address is valid. If the current cache miss address is valid, the list prefetch engine compares the current cache miss address and a list address. A list address represents an address in a list. A list describes an arbitrary sequence of prior cache miss addresses. The prefetch engine prefetches data according to the list, if there is a match between the current cache miss address and the list address.

97 MATHEMATICS AND COMPUTING↗