Search NASA⌕ Search

SEARCH · Search NASA

Results for “Accelerators, Computing”

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 1,765 records · Page 98

Considerations on the Use of Custom Accelerators for Big Data Analytics

Accelerators, including Graphic Processing Units (GPUs) for gen- eral purpose computation, many-core designs with wide vector units (e.g., Intel Phi), have become a common component of many high performance clusters. The appearance of more stable and reliable tools tools that can automatically convert code written in high-level specifications with annotations (such as C or C++) to hardware de- scription languages (High-Level Synthesis - HLS), is also setting the stage for a broader use of reconfigurable devices (e.g., Field Pro- grammable Gate Arrays - FPGAs) in high performance system for the implementation of custom accelerators, helped by the fact that new processors include advanced cache-coherent interconnects for these components. In this chapter, we briefly survey the status of the use of accelerators in high performance systems targeted at big data analytics applications. We argue that, although the progress in the use of accelerators for this class of applications has been sig- nificant, differently from scientific simulations there still are gaps to close. This is particularly true for the ”irregular” behaviors exhibited by no-SQL, graph databases. We focus our attention on the limits of HLS tools for data analytics and graph methods, and discuss a new architectural template that better fits the requirement of this class of applications. We validate the new architectural templates by mod- ifying the Graph Engine for Multithreaded System (GEMS) frame- work to support accelerators generated with such a methodology, and testing with queries coming from the Lehigh University Benchmark (LUBM). The architectural template enables better supporting the task and memory level parallelism present in graph methods by sup- porting a new control model and a enhanced memory interface. We show that out solution allows generating parallel accelerators, pro- viding speed ups with respect to conventional HLS flows. We finally draw conclusions and present a perspective on the use of reconfig- urable devices and Design Automation tools for data analytics.

Data Analytics↗

Transforming Energy through Computational Science: Computing for Clean Energy

This fact sheet discusses the opportunity space for NREL's computational science capabilities to address national clean energy objectives. Achieving a carbon-free power sector by 2035 as a step towards a decarbonized U.S. energy economy in 2050 will require major advances in power generation, autonomous energy systems, transportation, and buildings/communities. Development of integrated modeling approaches for complex energy systems will be essential for deployment. Success requires developments in optimization and control theory, complemented by machine learning (ML) and artificial intelligence (AI), all of which in turn need targeted investments in breadth and scale of computing. This document defines an opportunity space where: embracing computing can link established research and development (R&D) to a decarbonization agenda; pursuing emerging approaches can accelerate the pace of technology advancement across the portfolio; and leading by example could reduce the carbon footprint of computing worldwide.

advanced computing↗

A Taxonomy for Classification and Comparison of Dataflows for GNN Accelerators

Recently, Graph Neural Networks (GNNs) have received a lot of interest because of their success in learning representations from graph structured data. However, GNNs exhibit different compute and memory characteristics compared to traditional Deep Neural Networks (DNNs). Graph convolutions require feature aggregations from neighboring nodes (known as the aggregation phase), which leads to highly irregular data accesses. GNNs also have a very regular compute phase that can be broken down to matrix multiplications (known as the combination phase). All recently proposed GNN accelerators utilize different dataflows and microarchitecture optimizations for these two phases. Different communication strategies between the two phases have been also used. However, as more custom GNN accelerators are proposed, the harder it is to qualitatively classify them and quantitatively contrast them. In this work, we present a taxonomy to describe several diverse dataflows for running GNN inference on accelerators. This provides a structured way to describe and compare the design-space of GNN accelerators.

97 MATHEMATICS AND COMPUTING↗

Multiphase flow simulations of a moving fluidized bed regenerator in a carbon capture unit

To accelerate the commercialization and deployment of carbon capture technologies, computational fluid dynamics (CFD)-based tools may be used to model and analyze the performance of carbon capture devices. This work presents multiphase CFD-based flow simulations for the regeneration device responsible for extracting CO 2 from CO 2 -loaded sorbent particles before the particles are recycled. The use of solid particle sorbents in this design is a departure from previously reported systems, where aqueous sorbents are employed. Another new feature is the inclusion of a series of perforated plates along the regenerator height. The influence of these plates on sorbent distribution is examined for varying sorbent holdup, fluidizing gas velocity, and particle size. The residence time distribution of sorbents is also measured to classify the low regime as plug flow or well-mixed flow. The purpose of this work is to better understand the sorbent flow characteristics before reaction kinetics of CO 2 desorption can be implemented.

01 COAL, LIGNITE, AND PEAT↗

Bulky Phosphine Ligands Promote Palladium-Catalyzed Protodeboronation

The Suzuki-Miyaura cross-coupling reaction is plagued by protodeboronation, an undesirable side reaction with water that consumes the boronic acid derivatives required for the cross-coupling reaction. Meticulous mechanistic studies have previously established protodeboronation to be highly sensitive to the nature of the boronic reagent and reaction conditions. Particularly, the presence of bases, which are essential for the Suzuki-Miyaura coupling, is known to catalyze protodeboronation. However, protodeboronation catalyzed by palladium-phosphine complexes, the benchmark catalyst system for Suzuki-Miyaura cross-coupling, has been understudied compared to its base-catalyzed counterpart. Here, we demonstrate, using automated high-throughput experimentation, comprehensive computational mechanistic analyses and kinetic modeling, that protodeboronation is accelerated by palladium(II) complexes bound to bulky phosphine ligands. While sterically hindered ligands are typically used to facilitate difficult cross-couplings, these ligands can instead paradoxically impede cross-coupling product formation, requiring careful and judicious consideration when choosing ligands for Suzuki-Miyaura cross-couplings.

Ser, Cher Tian [University of Toronto, ON (Canada)↗

Full-dimensional global potential energy surfaces describing abstraction and exchange for the H + H{sub 2}S reaction

For the H + H{sub 2}S system, ∼34 000 data points are sampled over a large configuration space including both abstraction and exchange channels, and calculated at the level of explicitly correlated unrestricted coupled cluster method with singles, doubles, and perturbative triples excitations with the augmented correlation-consistent polarized triple zeta basis set (UCCSD(T)-F12a/aug-cc-pVTZ). The data set was fit using the newly proposed permutation invariant polynomial-neural network (PIP-NN) method with three different vectors as the input: two redundant sets of PIPs, one with the maximum order four (PES-I) and one with the maximum order three (PES-II), and nine non-redundant PIPs (PES-III). All these PESs show small fitting errors and essentially the same performance in representing the title system. Various kinetics and dynamical properties are calculated using the tunneling corrected transition state theory and quasi-classical trajectory, and compared with available experimental results. At a collision energy of 10 kcal/mol, both the H{sub 2} and SH products are found to be internally cold, with ∼20% of H{sub 2} at its first vibrational excited state, while SH is essentially a spectator. The angular distributions of the products are mainly in backward with considerable contributions from sideway direction. In addition, analytical partial derivatives of any PIP-NN PES with respect to the coordinates of atoms are derived by making use of the monomial symmetrization algorithm [Z. Xie and J. M. Bowman, J. Chem. Theory Comput. 6, 26–34 (2010)]. It can not only accelerate the evaluation of the derivatives, but also improve the energy convergence significantly.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

High-Fidelity CFD Simulation of Mixed Convection and Forced Convection in a Pebble Bed Test Reactor Core

The Hermes low-power [35-MW(thermal)] reactor will be built and operated by Kairos Power LLC (KP) to demonstrate its fluoride salt-cooled high-temperature reactor (FHR) technology. In the KP FHR, the reactor core is composed of randomly packed pebbles with TRISO fuel particles inside with FLiBe flow upward through the core acting as a coolant. Previous numerical and experimental studies have been limited to either a small-size bed or to a lack of detailed measurements for heat transfer. Here, to address the lack of high-fidelity heat transfer data in a real-size FHR core, in this study, we simulated a pebble bed core with 34 374 pebbles randomly packed, similar to the Hermes reactor's size. The core radius was 14 times that of the pebble diameter, while the core height was 45 times. In this work, we were particularly interested in a mixed convection regime, where buoyancy is important. Therefore, we performed several large-eddy simulations at different Reynolds numbers (160 to 1000) with gravitational force included. The spectral element computational fluid dynamics code NekRS with graphics processing unit acceleration was used for this study. The low-Mach number approximation was applied to address property changes in the FLiBe and to account for buoyancy. A pure hexahedral mesh with 60 million elements was generated by the Voronoi cell method. At the polynomial order of 5, the total degrees of freedom was 7.5 billion. The developed case in this work is the first of its kind in terms of size and complexity. The local numerical data across the domain were obtained and compared with empirical correlations. After examining the data, we found the following conclusions. For pressure drop, the Reger correlation predicted less than a 5% error. On the other hand, for heat transfer, the Wakao correlation outperformed the others. Based on our findings, we recommend the use of the Wakao correlation for the Nusselt number calculation, and for pressure drop, the KTA (Kerntechnischer Ausschuss) correclation, among the available experimental correlations. In conclusion, the Reger direct numerical simulation-driven correlation for pressure drops should also be considered, given its best agreement with our calculations.

Mixed Convection↗

The COMPRES/GSECARS gas-loading system for diamond anvil cells at the Advanced Photon Source

We have designed and constructed a new system for loading gases at high pressure into diamond anvil cells at pressures up to 200 MPa. The gases are used either as quasi-hydrostatic pressure media surrounding the sample or as the sample itself. The diamond cell is sealed using a clamping mechanism, which permits nearly any type of diamond anvil cell to be used. Online ruby fluorescence and video imaging systems allow in situ monitoring of the pressure and gasket deformation as the cell is sealed, resulting in a very high success rate in loading cells. The system includes interlocks and computer control that allow it to be safely and easily operated by visiting users at the Advanced Photon Source. We present preliminary X-ray diffraction data on volume compression of single-crystal magnesium oxide (MgO) in helium up to 110 GPa.

43 PARTICLE ACCELERATORS↗

Weakly interacting dark matter from the minimal walking technicolor

We study a weakly interacting dark matter particle motivated by minimal walking technicolor theories. Our WIMP is a mixture of a sterile state and a state with the charges of a standard model fourth family neutrino. We show that the model can give the right amount of dark matter over a range of the WIMP mass and mixing angle. We compute bounds on the model parameters from the current accelerator data including the oblique corrections to the precision electroweak parameters, as well as from cryogenic experiments, Super-Kamiokande and from the IceCube experiment. We show that consistent dark matter solutions exist which satisfy all current constraints. However, almost the entire parameter range of the model lies within the the combined reach of the next generation experiments.

79 ASTRONOMY AND ASTROPHYSICS↗

Synthesis of single-qutrit circuits from Clifford+𝑅 gates

Here, we present two deterministic compilation algorithms for single-qutrit unitaries with O ( log 1 / ɛ ) gate depth. Each algorithm selects a nearby approximation to the target unitary and then exactly synthesizes the approximation over the Clifford + R basis. The first algorithm exhaustively searches over the group; while the second algorithm searches only for Householder reflections. The exhaustive search algorithm yields an average R count of 2.193 ( 11 ) + 8.621 ( 7 ) log 10 ( 1 / ɛ ) , albeit with a time complexity of O ( ɛ − 4.4 ) . The Householder search algorithm results in a larger average R count of 3.20 ( 13 ) + 10.77 ( 3 ) log 10 ( 1 / ɛ ) at a reduced time complexity of O ( ɛ − 0.42 ) , greatly extending the reach in ɛ . These costs correspond asymptotically to 35% and 69% more non-Clifford gates compared with synthesizing the same unitary with two qubits. Such initial results are encouraging for using the R gate as the nontransversal gate for qutrit-based computation.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Enabling discovery data science through cross-facility workflows

Experimental and observational instruments for scientific research (such as light sources, genome sequencers, accelerators, telescopes and electron microscopes) increasingly require High Performance Computing (HPC) scale capabilities for data analysis and workflow processing. Next-generation instruments are being deployed with higher resolutions and faster data capture rates, creating a big data crunch that cannot be handled by modest institutional computing resources. Often these big data analysis pipelines also require near real-time computing and have higher resilience requirements than the simulation and modeling workloads more traditionally seen at HPC centers. While some facilities have enabled workflows to run at a single HPC facility, there is a growing need to integrate capabilities across HPC facilities to enable cross-facility workflows, either to provide resilience to an experiment, increase analysis throughput capabilities, or to better match a workflow to a particular architecture. In this paper we describe the barriers to executing complex data analysis workflows across HPC facilities and propose an architectural design pattern for enabling scientific discovery using cross-facility workflows that includes orchestration services, application programming interfaces (APIs), data access and co-scheduling.

Antypas, Katerina B.↗

ESnet/JLab FPGA Accelerated Transport

To increase the science rate for high data rates/volumes, Thomas Jefferson National Accelerator Facility (JLab) has partnered with Energy Sciences Network (ESnet) to define an edge to data center traffic shaping / steering transport capability featuring data event aware network shaping and forwarding. The keystone of this ESnet+JLab FPGA Accelerated Transport (EJFAT) is the joint development of an AI/ML directed dynamic compute work Load Balancer (LB) of UDP streamed data. The LB is a suite consisting of a Field Programmable Gate Array (FPGA) executing the dynamically configurable, low fixed latency LB data plane featuring real-time packet redirection and high throughput, and a control plane running on the FPGA host computer that monitors network and compute farm telemetry in order to make dynamic AI/ML guided decisions for destination compute host redirection/load balancing and destination resource provisioning. The LB provides for three-tier horizontal scaling across LB suites, core compute hosts, and CPUs within a host. The LB effectively provides seamless integration of edge/core computing to support direct experimental data processing for immediate use by JLab science programs and others such as the EIC as well as data centers of the future requiring high throughput and low latency for both hot and cooled data for both running experiment data acquisition systems and data center use cases.

97 MATHEMATICS AND COMPUTING↗

SU-F-J-21: Clinical Evaluation of Surface Scanning Systems in Different Treatment Locations

Purpose: To reduce imaging dose in fractionated IGRT, the ability of optical surface imaging systems (OSIS) to detect setup errors was tested. Therefore, clinical studies to evaluate for different treatment locations setup corrections derived by OSIS in comparison with x-ray image guidance in fractionated radiation therapy was performed. Methods: The setup correction accuracy of an OSIS system (AlignRT, VisionRT, London, UK) will be analysed for the 4 tumour locations Pelvis, Upper Abdomen, Thorax and Breast, 20 patients for each location in comparison to a different system (Sentinel, C-RAD, SE). For each patient, the setup corrections of the cone-beam computed tomography (CBCT) of an Elekta Versa HD linear accelerator (Elekta, Crawley, UK) is considered as gold-standard and then compared with those of the OSIS for the first ten fractions retrospectively. There were no clinical decisions made based on the surrogate system. For the OSIS, the reference surface is highly important as it represents the actual ground truth. It can be obtained either with the system itself or the surface structure delineated in the planning CT can be imported via DICOM interface. In this paper, the first results for the treatment region thorax are presented. The reference image modalities were compared. Results: Table 1 displays the difference between the setup corrections obtained with OSIS and CBCT in lateral (LAT), longitudinal (LNG) and vertical (VRT) direction for the DICOM reference image. While the median deviations are within a few millimeters, some outliers showed large deviations. Generally, the mean deviation as well as the spread was smallest in lateral and largest in vertical direction. Conclusion: Although the system allows fast, simple and non-invasive determination of setup corrections, it should be evaluated treatment region dependant. Therefore, the study is ongoing. The application of OSIS may help to reduce the imaging dose for the patient. We gratefully acknowledge the support by our colleagues from the Radiological University Clinic Heidelberg, where the study was performed. This work was funded by the Federal Ministry of Education and Research (BMBF) Germany, grant number 01IB13001B.

60 APPLIED LIFE SCIENCES↗

An organ and effective dose study of XVI and OBI cone‐beam CT systems

The main purpose of this work was to quantify patient organ doses from the two kilovoltage cone beam computed tomography (CBCT) systems currently available on medical linear accelerators, namely the X‐ray Volumetric Imager (XVI, Elekta Oncology Systems) and the On‐Board Imager (OBI, Varian Medical Systems). Organ dose measurements were performed using a fiber‐optic coupled (FOC) dosimetry system along with an adult male anthropomorphic phantom for three different clinically relevant scan sites: head, chest, and pelvis. The FOC dosimeter was previously characterized at diagnostic energies by Hyer et al. [Med Phys 2009;36(5):1711–16] and a total uncertainty of approximately 4% was found for in‐phantom dose measurements. All scans were performed using current manufacturer‐installed clinical protocols and appropriate bow‐tie filters. A comparison of image quality between these manufacturer‐installed protocols was also performed using a Catphan 440 image quality phantom. Results indicated that for the XVI, the dose to the lens of the eye (1.07 mGy) was highest in a head scan, thyroid dose (19.24 mGy) was highest in a chest scan, and gonad dose (29 mGy) was highest in a pelvis scan. For the OBI, brain dose (3.01 mGy) was highest in a head scan, breast dose (5.34 mGy) was highest in a chest scan, and gonad dose (34.61 mGy) was highest in a pelvis scan. Image quality measurements demonstrated that the OBI provided superior image quality for all protocols, with both better spatial resolution and low‐contrast detectability. The measured organ doses were also used to calculate a reference male effective dose to allow further comparison of the two machines and imaging protocols. The head, chest, and pelvis scans yielded effective doses of 0.04, 7.15, and 3.73 mSv for the XVI, and 0.12, 1.82, and 4.34 mSv for the OBI, respectively. PACS number: 87.57.uq

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Tool Support for Software Lookup Table Optimization

A number of scientific applications are performance-limited by expressions that repeatedly call costly elementary functions. Lookup table (LUT) optimization accelerates the evaluation of such functions by reusing previously computed results. LUT methods can speed up applications that tolerate an approximation of function results, thereby achieving a high level of fuzzy reuse. One problem with LUT optimization is the difficulty of controlling the tradeoff between performance and accuracy. The current practice of manual LUT optimization adds programming effort by requiring extensive experimentation to make this tradeoff, and such hand tuning can obfuscate algorithms. In this paper we describe a methodology and tool implementation to improve the application of software LUT optimization. Our Mesa tool implements source-to-source transformations for C or C++ code to automate the tedious and error-prone aspects of LUT generation such as domain profiling, error analysis, and code generation. We evaluate Mesa with five scientific applications. Our results show a performance improvement of 3.0× and 6.9× for two molecular biology algorithms, 1.4× for a molecular dynamics program, 2.1× to 2.8× for a neural network application, and 4.6× for a hydrology calculation. We find that Mesa enables LUT optimization with more control over accuracy and less effort than manual approaches.

Wilcox, Chris↗

Evolution of the SLATE linear algebra library

SLATE (Software for Linear Algebra Targeting Exascale) is a distributed, dense linear algebra library targeting both CPU-only and GPU-accelerated systems, developed over the course of the Exascale Computing Project (ECP). While it began with several documents setting out its initial design, significant design changes occurred throughout its development. In some cases, these were anticipated: an early version used a simple consistency flag that was later replaced with a full-featured consistency protocol. In other cases, performance limitations and software and hardware changes prompted a redesign. Sequential communication tasks were parallelized; host-to-host MPI calls were replaced with GPU device-to-device MPI calls; more advanced algorithms such as Communication Avoiding LU and the Random Butterfly Transform (RBT) were introduced. Early choices that turned out to be cumbersome, error prone, or inflexible have been replaced with simpler, more intuitive, or more flexible designs. Applications have been a driving force, prompting a lighter weight queue class, nonuniform tile sizes, and more flexible MPI process grids. Of paramount importance has been building a portable library that works across several different GPU architectures – AMD, Intel, and NVIDIA – while keeping a clean and maintainable codebase. Here we explore the evolving design choices and their effects, both in terms of performance and software sustainability.

Gates, Mark↗

A Prototype Wire Position Monitoring System

The Wire Position Monitoring System (WPM) will track changes in the transverse position of LCLS Beam Position Monitors (BPMs) to 1{micro}m over several weeks. This position information will be used between applications of beam based alignment to correct for changes in component alignment. The WPM system has several requirements. The sensor range must be large enough so that precision sensor positioning is not required. The resolution needs to be small enough so that the signal can be used to monitor motion to 1{micro}m. The system must be stable enough so that system drift does not mimic motion of the component being monitored. The WPM sensor assembly consists of two parts, the magnetic sensor and an integrated lock-in amplifier. The magnetic sensor picks up a signal from the alternating current in a stretched wire. The voltage v induced in the sensor is proportional to the wire displacement from the center of the sensor. The integrated lock-in amplifier provides a DC output whose magnitude is proportional to the AC signal from the magnetic sensor. The DC output is either read on a digital voltmeter or digitized locally and communicated over a computer interface.

43 PARTICLE ACCELERATORS↗

Biofidelic Digital Head Model Software

Researchers at Los Alamos have extensive capabilities in biofidelic modeling, simulation and software development for predicting traumatic brain injury (TBI). Since injury-causing experiments cannot be performed on live humans, and head dummies are incapable of predicting brain deformations/strain, our approach uses magnetic resonance imaging (MRI) data for the brain combined with advanced computational methods and substructure constitutive properties to identify harmful accelerations that may develop diffused axonal injury in the brain. The goal is to develop this capability to analyze the effectiveness of personal protective equipment (PPE) such as, helmets and the advanced energy-absorbing materials used in PPE to mitigate brain injury. Los Alamos is seeking a collaborative partner (through a cooperative research and development agreement) to develop this capability toward a potential technology that will have commercial viability.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗