Performance Portability in the Uintah Runtime System Through the Use of Kokkos.
Abstract not provided.
Sunderland, Daniel↗
SEARCH · Search NASA
Results for “Portable”
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.
Low-cost, Arduino-based, portable device for measurement of methane composition in biogas
Not provided.
Energy & Fuels↗
A Simple and Portable Multi-Channel Pyrometer allowing temperature measurements down to 800 K on the microsecond scale
The measurement of transient temperatures less than 1000 K for samples in laser-heated diamond anvil cells remains a challenge. Here we present a design and performance characteristics for a multi-channel pyrometer that works in the NIR from 1200 to 2000 nm. It has a relatively small footprint, is portable, only requires low voltage power supplies and can report temperatures down to 800 K on the millisecond scale or faster. A single data point without averaging can be acquired in 14 microseconds (sampling rate of 70'000 times/s). In conjunction with a diamond anvil cell, the system still works down to about 830 K. The pyrometer has been interfaced several times with the laser heating system at the High Pressure Collaborative Access Team (HPCAT) at the Advanced Photon Source in Chicago with good results.
Montgomery, Jeffrey M.↗
High-Performance Molecular Dynamics Simulation for Biological and Materials Sciences: Challenges of Performance Portability, Dallas, TX, November 2018.
Highly-optimized parallel molecular dynamics programs have allowed researchers to achieve ground-breaking results in biological and materials sciences. This type of performance has come at the expense of portability: a significant effort is required for performance optimization on each new architecture. Using a metric that emphasizes speedup, we assess key accelerating programming components of four different best-performing molecular dynamics programs - GROMACS, NAMD, LAMMPS and CP2K-- each having a particular scope of application, for contribution to performance and for portability. We use builds with and without these components, tested on HPC systems. We also analyze the code-bases to determine compliance with portability recommendations. We find that for all four programs, the contributions of the non-portable components to speed are essential to the programs' performances; without them we see a reduction in time-to-solution of a magnitude that is insufferable to domain scientists. This characterizes the performance efficiency that must be approached for good performance portability on a programmatic level, suggesting solutions to this difficult problem, which should come from developers, industry and funding institutions, and possibly new research in programming languages.
Computer Science↗
Demonstrating GPU code portability and scalability for radiative heat transfer computations
High performance computing frameworks utilizing CPUs, Nvidia GPUs, and/or Intel Xeon Phis necessitate portable and scalable solutions for application developers. Nvidia GPUs in particular present numerous portability challenges with a different programming model, additional memory hierarchies, and partitioned execution units among streaming multiprocessors. This work presents modifications to the Uintah asynchronous many-task runtime and the Kokkos portability library to enable one single codebase for complex multiphysics applications to run across different architectures. Scalability and performance results are shown on multiple architectures for a globally coupled radiation heat transfer simulation, ranging from a single node to 16,384 Titan compute nodes.
Computer Science↗
A portable cryostat for the cold transfer of polarized solid HD targets: HDice-I
We developed a device with moveable liquid nitrogen and liquid helium volumes that is capable of reaching over 2 m into the coldest regions of a cryostat or dilution refrigerator and reliably extracting or installing a target of solid, polarized hydrogen deuteride (HD). This Transfer Cryostat incorporates a cylindrical neodymium rare-earth magnet that is configured as a Halbach dipole, which is maintained at 77 K and produces a 0.1 T field around the HD target. Multiple layers provide a hermetic 77 K-shield as the device is used to maintain a target at 2 K during a transfer between cryostats. Our tests with frozen-spin HD show very little polarization loss for either H (-1±2%, relative) or D (0±3%, relative) over typical transfer periods. Multiple target transfers with this apparatus have shown an overall reliability of about 95% per transfer, which is a significant improvement over earlier versions of the device.
42 ENGINEERING↗
JACC.shared: Leveraging HPC Metaprogramming and Performance Portability for Computations That Use Shared Memory GPUs
In this work, we present JACC.shared, a new feature of Julia for ACCelerators (JACC), which is the performanceportable and metaprogramming model of the just-in-time and LLVM-based Julia language. This new feature allows JACC applications to leverage the high-performance computing (HPC) capabilities of high-bandwidth, on-chip GPU memory. Historically, exploiting high-bandwidth, shared-memory GPUs has not been a priority for high-level programming solutions. JACC.shared covers that gap for the first time, thereby providing a highlevel, portable, and easy-to-use solution for programmers to exploit this memory and supporting all current major accelerator architectures. Well-known HPC and AI workloads, such as multi/hyperspectral imaging and AI convolutions, have been used to evaluate JACC.shared on two exascale GPU architectures hosted by some of the most powerful US Department of Energy supercomputers: Perlmutter (NVIDIA A100) and Frontier (AMD MI250X). The performance evaluation reports speedup of up to 3.5× by adding only one line of code to the base codes, thus providing important accelerators in a simple, portable, and transparent way and elevating the programming productivity and performance-portability capabilities for Julia/JACC HPC, AI, and scientific applications.
Valero Lara, Pedro [ORNL] (ORCID:0000000214794310)↗
Development and implementation of a portable grating interferometer system as a standard tool for testing optics at the Advanced Photon Source beamline 1-BM
We developed a portable X-ray grating interferometer setup as a standard tool for testing optics at the Advanced Photon Source (APS) beamline 1-BM. The interferometer can be operated in phase-stepping, Moiré, or single-grating harmonic imaging mode with 1-D or 2-D gratings. All of the interferometer motions are motorized; hence, it is much easier and quicker to switch between the different modes of operation. A novel aspect of this new instrument is its designed portability. While the setup is designed to be primarily used as a standard tool for testing optics at 1-BM, it could be potentially deployed at other APS beamlines for beam coherence and wavefront characterization or imaging. Herein, the design of the interferometer system is described in detail and coherence measurements obtained at the APS 34-ID-E beamline are presented. The coherence was probed in two directions using a 2-D checkerboard, a linear, and a circular grating at X-ray energies of 8 keV, 11 keV, and 18 keV.
47 OTHER INSTRUMENTATION↗
Design and implementation of a portable diagnostic system for Thomson scattering and optical emission spectroscopy measurements
A diagnostic system, which has a design goal of high-portability, has been designed at Oak Ridge National Laboratory (ORNL). This project aims at providing measurements of key plasma parameters (ne, Te, ni, Ti) for fusion-relevant devices, utilizing Thomson scattering (TS) and optical emission spectroscopy (OES). The innovative design employs mostly commercial off-the-shelf instrumentation and a traveling team of researchers to conduct measurements at various magnetic-confinement plasma devices. The TS diagnostic uses a Quantel Q-smart 1500 Nd:YAG laser with a 2ω harmonic generator to produce up to 850 mJ of 532 nm laser pulses at 10 Hz. Collection optics placed at the detection port consists of an 11 × 3 optical fiber bundle, where the TS diagnostic uses an 11 × 1 subset array of the fibers, the OES diagnostic uses another 11 fibers, and the remaining fibers are available to the host institution. The detection system is comprised of two separate IsoPlane-320 spectrometers with triple-grating turrets of various line spacing and two PI-MAX 4 intensified CCD detectors, used simultaneously to measure a broad range of ion, impurity, and electron parameters. The self-contained diagnostic package also includes a data processing and storage system. The design and initial implementation of the TS-OES diagnostic system are described. The experiments from the proof-of-principle operation of the portable package on a high density (∼2.5 × 1022 m−3) and low-temperature (∼5 eV) electrothermal arc source at ORNL are also discussed.
Kafle, N. (ORCID:000000028144787X)↗
Portability for GPU-accelerated molecular docking applications for cloud and HPC: can portable compiler directives provide performance across all platforms?
High-throughput structure-based screening of drug-like molecules has become a common tool in biomedical research. Recently, acceleration with graphics processing units (GPUs) has provided a large performance boost for molecular docking programs. Both cloud and high-performance computing (HPC) resources have been used for large screens with molecular docking programs; while NVIDIA GPUs have dominated cloud and HPC resources, new vendors such as AMD and Intel are now entering the field, creating the problem of software portability across different GPUs. Ideally, software productivity could be maximized with portable programming models that are able to maintain high performance across architectures. While in many cases compiler directives have been used as an easy way to offload parallel regions of a CPU-based program to a GPU accelerator, they may also be an attractive programming model for providing portability across different GPU vendors, in which case the porting process may proceed in the reverse direction: from low-level, architecture-specific code to higher-level directive-based abstractions. MiniMDock is a new mini-application (miniapp) designed to capture the essential computational kernels found in molecular docking calculations, such as are used in phar-maceutical drug discovery efforts, in order to test different solutions for porting across GPU architectures. Here we extend MiniMDock to GPU offloading with OpenMP directives, and compare to performance of kernels using CUDA and HIP on NVIDIA and AMD GPUs, respectively, as well as across different compilers, exploring performance bottlenecks. We document this reverse-porting process, from highly optimized device code to a higher-level version using directives, compare code structure, and describe barriers that were overcome in this effort.
Thavappiragasam, Mathialakan↗
Fuel Cells for Portable Power: 1. Introduction to DMFCs; 2. Advanced Materials and Concepts for Portable Power Fuel Cells [Slides]
Thanks to generally less stringent cost constraints, portable power fuel cells, the direct methanol fuel cell (DMFC) in particular, promise earlier market penetration than higher power polymer electrolyte fuel cells (PEFCs) for the automotive and stationary applications. However, a large-scale commercialization of DMFC-based power systems beyond niche applications already targeted by developers will depend on improvements to fuel cell performance and performance durability as well as on the reduction in cost, especially of the portable systems on the higher end of the power spectrum (100-250 W). In this part of the webinar, we will focus on the development of advanced materials (catalysts, membranes, electrode structures, and membrane electrode assemblies) and fuel cell operating concepts capable of fulfilling two key targets for portable power systems: the system cost of $5/W and overall fuel conversion efficiency of 2.0-2.5 kWh/L. Presented research will concentrate on the development of new methanol oxidation catalysts, hydrocarbon membranes with reduced methanol crossover, and improvements to component durability. Time permitted, we will also present a few highlights from the development of electrocatalysts for the oxidation of two alternative fuels for the direct-feed fuel cells: ethanol and dimethyl ether.
10 SYNTHETIC FUELS↗
Using a Customized Portable Deepwater Portable Electrofisher to Assess Larval Lamprey Populations in Irrigation Canals
Larval lamprey densities were quantified at two large-scale water diversions facilities located on the Yakima River near the city of Yakima, Washington. Water diversions are very common in the Yakima River Basin and provide water to irrigate a diverse range of valuable cropland. Out-migrating larval Pacific Lamprey are vulnerable to entrainment through fish protection screens during the withdrawal period and desiccation or predation during dewatering at the end of the season. In order to determine the impact to lamprey populations which inhabit these regions a portable deepwater electroshocking system was utilized to determine lamprey densities which could otherwise not be determined due to water depth. Surveys were conducted in the fall during the dewatering period in 2015 and 2017 and larval lamprey densities ranged from 0.8 to 8.8 fish m-2 at the Sunnyside Diversion Facility to 4.2 to 6.5 fish m-2 at the Wapato Diversion Facility. The estimates for the total lamprey numbers at each of the region using a portable deepwater electrofishing platform (PDEP) was very similar to those estimates using backpack electrofishing post-dewatering; the PDEP method was 12- 36% higher where comparisons were available. During 2017, the estimated lamprey entrainment at the Sunnyside Diversion increased by 484% from August to November. Lamprey lengths were also determined and ranged from 50 to 130 mm. Our results indicate that the use of the deepwater shocking system was very safe and effective at determining larval lamprey densities at hard to sample regions which are present near irrigation facilities.
larval lamprey, electrofishing, water diversion↗
Implementation of a portable diagnostic system for Thomson scattering measurements on an electrothermal arc source
To fulfill the increasing needs of diagnostic support for researchers in plasma technology, a portable diagnostic package (PDP) equipped for both laser Thomson scattering (TS) and optical emission spectroscopy has been designed and constructed at Oak Ridge National Laboratory (ORNL), aiming to measure the temperature and number density of electrons and temperatures of ions in plasma devices. The PDP has been initially implemented on a high density and low temperature electrothermal arc source (ET-arc) at ORNL to test its TS capability. TS from the plasmas in the ET-arc has been obtained using the PDP. The electron temperature and number density were determined from TS spectra. These results were then compared to measurements from previous studies on the ET-arc. The TS diagnostic measured 0.8 ± 0.1, 1.3 ± 0.2, and 0.7 ± 0.1 eV and (4.4 ± 0.5) × 1021, (5.9 ± 0.7) × 1021, and (4.3 ± 0.5) x 1021 m-3, respectively, from three lines of sight that transect the plasma column.
He, Z. (ORCID:0000000183159882)↗
Portable diagnostic package for Thomson scattering and optical emission spectroscopy on Princeton field-reversed configuration 2 (PFRC 2)
An Advanced Research Projects Agency-Energy funded diagnostic system has been deployed to the Princeton field-reversed configuration 2 (PFRC-2) device, located at Princeton Plasma Physics Laboratory. The Portable Diagnostic Package (PDP), designed at Oak Ridge National Laboratory, allows for the measurement of Thomson Scattering (TS) for electron density and temperature and Optical Emission Spectroscopy (OES) for ion temperature, impurity density, and ion velocity. A tunable spectrometer on the PDP with three gratings provides the flexibility to measure low (1 eV) and high (1000 eV) electron temperature ranges from TS. Additionally, using a second spectrometer, the OES diagnostic can survey light emission from various ion excitation levels for wide wavelength ranges. The electron density (<2 × 10 19 m –3 ) of plasmas generated in PFRC-2 has been below the PDP TS discrimination threshold, which has made TS signal detection challenging against a high-background of laser stray light. The laser stray light was iteratively reduced by making modifications to the entrance and exit geometry on PFRC-2. Rayleigh scattering experiments on PFRC have yielded the TS discrimination sensitivity to be >1 × 10 20 m –3 for the PDP. A recently implemented narrow-band notch spectral filter that masks the second harmonic 532 nm Nd:YAG laser wavelength has increased the system’s TS light discrimination sensitivity 65 times compared to the instance when the notch filter was not implemented. The hardware implementation including design changes to the flight tubes and Brewster windows will be discussed, along with results from Rayleigh and rotational Raman scattering sensitivity analyses, which were used to establish a quantitative figure of merit on the system performance. Further, the Raman scattering calibration with the notch filter has improved the PDP electron density threshold to 1 ± 0.5 × 10 18 m –3 .
47 OTHER INSTRUMENTATION↗
A portable and monoenergetic 24 keV neutron source based on 124Sb-9Be photoneutrons and an iron filter
A portable monoenergetic 24 keV neutron source based on the 124Sb-9Be photoneutron reaction and an iron filter has been constructed and characterized. The coincidence of the neutron energy from SbBe and the low interaction cross-section with iron (mean free path up to 29 cm) makes pure iron specially suited to shield against gamma rays from 124Sb decays while letting through the neutrons. To increase the 124Sb activity and thus the neutron flux, a >1 GBq 124Sb source was produced by irradiating a natural Sb metal pellet with a high flux of thermal neutrons in a nuclear reactor. The design of the source shielding structure makes for easy transportation and deployment. A hydrogen gas proportional counter is used to characterize the neutrons emitted by the source and a NaI detector is used for gamma background characterization. At the exit opening of the neutron beam, the characterization determined the neutron flux in the energy range 20–25 keV to be 6.00±0.30 neutrons per cm2 per second and the total gamma flux to be 245±8 gammas per cm2 per second (numbers scaled to 1 GBq activity of the 124Sb source). A liquid scintillator detector is demonstrated to be sensitive to neutrons with incident kinetic energies from 8 to 17 keV, so it can be paired with the source as a backing detector for neutron scattering calibration experiments. This photoneutron source provides a good tool for in-situ low energy nuclear recoil calibration for dark matter experiments and coherent elastic neutrino-nucleus scattering experiments.
Dark Matter detectors (WIMPs, axions, etc.)↗
JACC: Leveraging HPC Meta-Programming and Performance Portability with the Just-in-Time and LLVM-based Julia Language
We present JACC (Julia for Accelerators), the first high-level, and performance-portable model for the just-in-time and LLVM-based Julia language. JACC provides a unified and lightweight front end across different back ends available in Julia, enabling the same Julia code to run efficiently on many HPC CPU and GPU targets. We evaluated the performance of JACC for common HPC kernels as well as for the most computationally demanding kernels used in applications, HPCCG, a supercomputing benchmark test for sparse domains, and HARVEY, a blood flow simulator to assist in the diagnosis and treatment of patients suffering from vascular diseases. We carried out the performance analysis on the most advanced US DOE supercomputers: Aurora, Frontier, and Perlmutter. Overall, we show that JACC has a negligible overhead versus vendor-specific solutions, reporting GPU speedups with no extra cost to programmability.
Valero-Lara, Pedro↗
Towards Achieving Performance Portability Using Directives for Accelerators
In this paper we explore the performance portability of directives provided by OpenMP 4 and OpenACC to program various types of node architectures with attached accelerators, both self-hosted multicore and offload multicore/GPU. Our goal is to examine how successful OpenACC and the newer offload features of OpenMP 4.5 are for moving codes between architectures, how much tuning might be required and what lessons we can learn from this experience. To do this, we use examples of algorithms with varying computational intensities for our evaluation, as both compute and data access efficiency are important considerations for overall application performance. We implement these kernels using various methods provided by newer OpenACC and OpenMP implementations, and we evaluate their performance on various platforms including both X86_64 with attached NVIDIA GPUs, self-hosted Intel Xeon Phi KNL, as well as an X86_64 host system with Intel Xeon Phi coprocessors. In this paper, we explain what factors affected the performance portability such as how to pick the right programming model, its programming style, its availability on different platforms, and how well compilers can optimize and target to multiple platforms.
97 MATHEMATICS AND COMPUTING↗
Repeatability, accuracy, portability, and errors of the portable alignment gyrocompass system
Portable alignment gyrocompass system accuracy, repeatability, portability, and errors
Mihm, J. J.↗