Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel, scaling”

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 1,369 records · Page 76
All sourcesNASA NTRSDOE OSTI
DOE OSTI2022

Phase Field Dislocation Dynamics (PFDD) version 2.x

This disclosure is for version 2.x of a mesoscale model called Phase Field Dislocation Dynamics (PFDD). PFDD is used for investigating deformation in nanoscale (grain sizes of ~300 nm and less) materials, such as metals and alloys. This approach models the motion and interaction of individual defects, namely dislocations, in the material using scalar-valued phase field variables, also called order parameters. The system is evolved through energy minimization thus the model calculates the total energy density in terms of the phase field variables. The energy minimization is completed using the Ginzburg-Landau equation, and is implemented with explicit time integration. The total system energy can be comprised of several terms, including the strain energy (which describes dislocation-dislocation interactions), the energy due to an applied stress (dislocation interactions with the applied stress), and a core/lattice (perfect dislocations) or generalized stacking fault (partial dislocations) energy (described the dislocation core structure). The latter term in particular may vary based on the crystal structure being modeled and is typically informed using lower length scale (e.g., atomistic) approaches, although no such (atomistic) calculations are completed within the PFDD algorithm. This basic formulation was previously reviewed by Los Alamos National Laboratory and released under license number C17113. This previously reviewed version we will henceforth refer to as PFDD v1.0. PFDD v1.0 consisted of 2 codes (one parallel and one serial) plus input files, all written in the C language. This new disclosure is addressing the next versions of the PFDD, versions 2.x. There have been several enhancements of PFDD v1.0, which are described here and included in the attached code, which we will refer to as PFDD v2.0. There are also several new features described here that are either planned or already in process and are expected to be subsequent releases, i.e., v2.1, v2.2, ...v2.x.

Hunter, Abigail↗
DOE OSTI2018

Power Take-off System for Marine Renewable Devices

The Power Take-off (PTO) System for Marine Renewable Devices (Project) envisioned the design, manufacture and testing of a generator that would ultimately be suitable for fully flooded seawater operation. This would require a generator design and bearing system capable of operating in the presence of seawater without corrosion. The original generator concept design focused on a switched reluctance electromagnetic design for the generator with the understanding that the switched reluctance design would be more robust and tolerant of a seawater medium. A trade study was conducted to prove that hypothesis, but the results of the study indicated that there was no benefit to adopting a switched reluctance design over a more conventional synchronous permanent magnet design. In addition, the novelty of the switched reluctance design added more technical risk to the Project than was warranted. ORPC’s cost of energy analyses pointed to reliability as being the most critical factor for achieving Project systems performance advancement (SPA) metrics. Consequently, the choice of a permanent magnet (PM) machine design over a potentially less reliable switched reluctance design was necessary to achieve the Project results. As a result, ORPC determined that a PM design was a more appropriate design to achieve the Project goals and ORPC selected a highly qualified PM generator firm for the design, manufacture and factory testing of the generator. The factory acceptance test (FAT) of the generator took place on March 20-22, 2017 and all tests were successfully passed. Developing a design for a driveline as part of the PTO system required a full understanding of the loads and operating conditions expected. Detailed analyses for developing turbine loads and subsequent definition of bearing specifications were performed using various analytical methods. Two candidate bearing designs were proposed as part of the Project. After initial design work had been completed, ORPC designed a representative full-scale driveline to validate the detailed design of the advanced driveline components. Full-scale components were manufactured, instrumented and tested, simulating actual system parameters. This full-scale bearing test was conducted at the University of Maine Advanced Structures and Composites Laboratory. The purpose of this test was to characterize the frictional losses associated with bearings and test the suitability and functionality of some commercially available shaft couplings and expansion bushings in cross-flow turbine drivelines. In parallel with component testing and dynamometer testing of the PTO system, ORPC performed design work to demonstrate how the advanced PTO system will be integrated with all ORPC turbine generator units (TGUs). The work would focus primarily on the TidGen® TGU and assess how the advanced PTO would be integrated into TidGen® 001 or other units, then field-deployed and tested in an open water environment under actual operational conditions. In addition to the benefits provided to the MHK industry in general, and ORPC in particular, the Project provided benefits to the general public, including the following: • Increased general awareness of the potential for generation of electricity from ocean energy resources • Demonstrated continuing advancements in the commercialization of MHK technologies, making them more likely to contribute to the electricity generation supply in America in the coming years • Continued awareness of the local economic benefits from the MHK industry through job creation and local spending • Confirmation of a path to improvements in availability, power-to-weight ratio (PWR) and reduction in levelized cost of energy (LCOE). All Project tasks were completed, and the Project objectives were accomplished. All project milestones and deliverables were met. The designs, methodologies, practices, testing, data, analysis and lessons learned from the Project are a step forward in the development of the U.S. MHK industry and provide a sound basis for ultimate commercialization of ORPC’s power systems. Based on the what was learned and demonstrated during the Project, ORPC has determined that the ORPC power systems have the potential of accelerating the commercialization of tidal and ocean current MHK power systems.

16 TIDAL AND WAVE POWER↗
DOE OSTI2020

10 MWE CDCL Large Pilot Plant – pre-FEED Study

Coal Direct Chemical Looping (CDCL) is an advanced oxy-combustion technology that has the potential to substantially reduce the energy penalty and the cost of electricity (COE) for coal-fired power generation with CO₂ capture. The Babcock & Wilcox Company (B&W) and The Ohio State University (OSU) have been collaborating on the development of an iron oxide oxygen-carrier based chemical looping technology for clean power generation with inherent carbon capture. In this process, coal is dried and pulverized prior to being transported into a moving-bed reducer. In the reducer, coal reacts with the oxygen-carrier particles, forming combustion byproducts, predominantly CO₂ and H₂O, while reducing the iron oxide oxidation state from Fe₂O₃ to a mixture of FeO and Fe. The reduced state particles are then transported to a combustor reactor and re-oxidized with air. Following the oxidation, the oxygen-carrier particles are regenerated, and a large amount of heat is released for steam production. The produced steam is sent to a turbine for electricity generation. Meanwhile, the CO₂–rich stream leaving the reducer is cooled, cleaned, and compressed for subsequent pipeline transportation and sequestration. By combining air separation and fuel conversion into a single system, the CDCL technology enables the intensification of oxy-combustion processes by eliminating the energy and cost intensive cryogenic air separation unit and thereby results in higher overall plant efficiencies and lower COE’s. The use of a moving-bed reducer results in high conversions of volatile hydrocarbons and high CO₂ purity, which reduces the cost of downstream CO₂ purification for sequestration or utilization. The Babcock & Wilcox Company in collaboration with The Ohio State University, Johnson & Matthey, The Electric Power Research Institute, and Dover Light & Power completed a Preliminary Front-End Engineering and Design (Pre-FEED) study of a 10 MWe coal-direct chemical looping (CDCL) pilot plant. The planned system is a modular 10 MWe CDCL large pilot facility consisting in 4 modules of 2.5 MWe each, working in parallel and to be hosted within the current structure at the City of Dover’s Municipal Power Plant. The CDCL system can achieve auto-thermal operation and includes a sub-critical steam cycle for power generation. The pilot system was designed to demonstrate full commercial operation at a reduced scale. The coal distribution per plan area is a representative slice of larger commercial arrangements. The system includes a CO₂ recycle system, but it does not include a compression system. The large pilot includes all the environmental control equipment and oxygen carrier and ash handling systems. As part of the project, the Team performed laboratory testing and carried out multiple pilot test campaigns to obtain design and performance information at the 250 kWth CDCL pilot facility at the Babcock & Wilcox Company’s Research Center. The Team demonstrated sustained operations at designed coal inputs, high coal conversion, high CO₂ purity, heat generation on the combustor, low carbon carryover between reactors and low particle attrition. Emissions generated in the reducer reactor were identified as SO₂ and NO x . The commercial manufacturing cost of oxygen carrier particle was evaluated by JM. A particle manufacturing report was generated and submitted to the DOE. Based on the results from the pilot tests and the pre-FEED design efforts, a techno-economic analysis was performed. The study shows that the CDCL process is a promising carbon-friendly technology capable of producing electricity with high efficiency. The estimated COE of the supercritical CDCL plant is $83.3 / MWh, which meets DOE’s target of less than 30% increase in COE when compared to a supercritical PC plant without CO₂ capture. This is the lowest among the existing carbon capture technologies (post-combustion and oxy-combustion) for fossil fuel power plant. CDCL is evaluated to be the most promising technology for carbon capture from the economic aspect.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2024

Cholla-MHD: An Exascale-capable Magnetohydrodynamic Extension to the Cholla Astrophysical Simulation Code

Abstract We present an extension of the massively parallel, GPU native, astrophysical hydrodynamics code Cholla to magnetohydrodynamics (MHD). Cholla solves the ideal MHD equations in their Eulerian form on a static Cartesian mesh utilizing the Van Leer + constrained transport integrator, the HLLD Riemann solver, and reconstruction methods at second and third order. Cholla’s MHD module can perform ≈260 million cell updates per GPU-second on an NVIDIA A100 while using the HLLD Riemann solver and second order reconstruction. The inherently parallel nature of GPUs combined with increased memory in new hardware allows Cholla’s MHD module to perform simulations with resolutions ∼500 3 cells on a single high-end GPU (e.g., an NVIDIA A100 with 80 GB of memory). We employ GPU direct Message Passing Interface to attain excellent weak scaling on the exascale supercomputer Frontier, while using 74,088 GPUs and simulating a total grid size of over 7.2 trillion cells. A suite of test problems highlights the accuracy of Cholla’s MHD module and demonstrates that zero magnetic divergence in solutions is maintained to round off error. We also present new testing and CI tools using GoogleTest, GitHub Actions, and Jenkins that have made development more robust and accurate and ensure reliability in the future.

Astronomy & Astrophysics↗
DOE OSTI2019

Panel Session 11: Featured Country Canada - Remediation of Contaminated Sites

Canada has several unique sites requiring environmental remediation of radiological contaminants. This panel was designed to outline the remediation of contaminated sites across Canada from several perspectives. Remote legacy uranium mining operations, small scale residential locations, and large impacted soil sites all pose different challenges, however there are some common solutions and keys to success. This panel addressed several of those challenges and provided lessons learned on how to successfully plan and execute large scale environmental remediation in both remote and residential locations. This panel focused on the cleaning up several contaminated sites in Canada and monitoring the long-term performance of sites that have already been remediated. Panelists addressed the technical and social challenges associated with various cleanup initiatives North of the 49. Parallel and the solutions that have been adopted to enable these projects to move forward. The panel offered the perspectives of owners, contractors and regulators on current projects and how they are leveraging past experience and learning from others to ensure future undertakings are successful. Presentations proposed include: 1 - Regulating Environmental Remediation in Canada (Michael Rinker, Canadian Nuclear Safety Commission (Canada)); 2 - Integration of Local Knowledge and Engagement/Capacity Building Legacy Uranium Mine Sites (Joe Muldoon, Saskatchewan Research Council (Canada)); 3 - The Path to Institutional Control - Safe, Secure and Stable (R. Liam Mooney, Cameco Corporation (Canada)); 4 - Port Hope Area Initiative: Project Challenges (Scott Parnell, Canadian Nuclear Laboratories (Canada))

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2023

gRNA-SeqRET: a universal tool for targeted and genome-scale gRNA design and sequence extraction for prokaryotes and eukaryotes

High-throughput genetic screening is frequently employed to rapidly associate gene with phenotype and establish sequence-function relationships. With the advent of CRISPR technology, and the ability to functionally interrogate previously genetically recalcitrant organisms, non-model organisms can be investigated using pooled guide RNA (gRNA) libraries and sequencing-based assays to quantitatively assess fitness of every targeted locus in parallel. To aid the construction of pooled gRNA assemblies, we have developed an in silico design workflow for gRNA selection using the gRNA Sequence Region Extraction Tool (gRNA-SeqRET). Built upon the previously developed CCTop, gRNA-SeqRET enables automated, scalable design of gRNA libraries that target user-specified regions or whole genomes of any prokaryote or eukaryote. Additionally, gRNA-SeqRET automates the bulk extraction of any regions of sequence relative to genes or other features, aiding in the design of homology arms for insertion or deletion constructs. We also assess in silico the application of a designed gRNA library to other closely related genomes and demonstrate that for very closely related organisms Average Nucleotide Identity (ANI) > 95% a large fraction of the library may be of relevance. The gRNA-SeqRET web application pipeline can be accessed at https://grna.jgi.doe.gov. The source code is comprised of freely available software tools and customized Python scripts, and is available at https://bitbucket.org/berkeleylab/grnadesigner/src/master/ under a modified BSD open-source license (https://bitbucket.org/berkeleylab/grnadesigner).

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2016

Large-scale large eddy simulation of nuclear reactor flows: Issues and perspectives

Numerical simulation has been an intrinsic part of nuclear engineering research since its inception. In recent years a transition is occurring toward predictive, first-principle-based tools such as computational fluid dynamics. Even with the advent of petascale computing, however, such tools still have significant limitations. In the present work some of these issues, and in particular the presence of massive multiscale separation, are discussed, as well as some of the research conducted to mitigate them. Petascale simulations at high fidelity (large eddy simulation/direct numerical simulation) were conducted with the massively parallel spectral element code Nek5000 on a series of representative problems. These simulations shed light on the requirements of several types of simulation: (1) axial flow around fuel rods, with particular attention to wall effects; (2) natural convection in the primary vessel; and (3) flow in a rod bundle in the presence of spacing devices. Finally, the focus of the work presented here is on the lessons learned and the requirements to perform these simulations at exascale. Additional physical insight gained from these simulations is also emphasized.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2023

Lime slurry treatment of soils developing on abandoned coal mine spoil: Linking contaminant transport from the micrometer to pedon-scale

Historical and abandoned coal mine spoil continues to generate acid- and metal(loid)-rich porewaters and represents a geographically large and diffuse non-point source of contamination to local watersheds. A potentially inexpensive approach to treat these materials and soils developing on them is through the application of lime slurries, to neutralize acidity and encourage the (co)precipitation of metal(loids) with Fe(III)-(oxy)hydroxides, and potentially other metal-oxides and/or Ca-bearing phases. Here, the efficacy of this approach was evaluated through parallel field application and laboratory-based flow-through column experiments. The field site is Huff Run sub-watershed 25 located in Tuscarawas County, Ohio and was chosen in part because it was previously classified as one of the most highly AMD-impacted sub-watersheds in the region. Two locations with historical spoil were chosen and suction lysimeters were installed at 25 and 75 cm depth to monitor porewater composition on two sides at the base of each pile. Half of each slope received seven lime slurry treatments from June through October of 2017. A suite of aqueous (ICP-OES, IC, and TOC-L) and solid phase geochemical and mineralogical approaches (quantitative SEM-EDS and synchrotron μ-XRF) were used to determine how composition, texture, morphology, and spatial distribution of mineral coatings differ in pre- and post-lime treated soils, and how that impacts the distribution and transport of trace metal(loid)s. Mine spoil porewater at site 1 was slightly less alkaline (pH ranging from 7.04 to 7.37) than at site 2 (ranging from 7.55 to 7.71), and average electrical conductivity values at site 1 (316–405 μS cm —1 ) were slightly lower than at site 2 (358–464 μS cm —1 ), although differences between the sites were not significant. Porewater pH and electrical conductivity in all lysimeters decreased over the course of the field season but there was no obvious response to lime treatment at either site or any depth. At site 1, both treatment and depth were significant factors affecting Ca, K, Ni, SO 4 2— , and DOC concentrations while only treatment effects were significant for dissolved Al and Cu (p < 0.05). For all soils, there were no trends in metal concentration observed over time although DOC and SO 4 2— decreased over the field season. Pedon-scale changes in metal porewater concentrations in response to treatment were linked to micrometer-scale changes in mineral surface coatings; specifically, higher concentrations of Ca, Fe, Mn, and Zn were observed in the coatings and no changes were observed in Fe redox speciation, whereas total S decreased likely due to oxidation of S in coal fragments. In contrast to the field experiment, the column experiments exhibited a much greater response in effluent composition with respect to lime treatment. The untreated columns had approximately an order of magnitude more H + leached over the course of the experiment (p < 0.001) and resulted in greater Ca, Al, Cu, and DIC leached and less Mn, Zn, and SO 4 2— . Soils treated with the lime slurry in the column experiments exhibited larger and thicker secondary Fe-coatings, including the addition of Fe-sulfates. Despite clear trends in the laboratory-based column experiment where the lime-to-soil ratio was higher, the effects were either muted or undetected in the field pilot project, suggesting that a higher application rate of lime in the field is needed to achieve a similar effect. This work provides evidence that a less alkaline lime slurry could be a practical and inexpensive method of treating coal mine spoil-impacted soils and represents an important step in linking laboratory-based remediation studies to implemented field-based studies.

58 GEOSCIENCES↗
DOE OSTI2021

Parallel I/O Evaluation Techniques and Emerging HPC Workloads: A Perspective

Emerging workloads such as artificial intelligence, big data analytics and complex multi-step workflows alongside future exascale applications are anticipated future HPC workloads, which will result in a more diverse I/O system workload and even less predictable I/O behavior and access patterns. Along with the ever increasing gap between the compute and storage performance capabilities, the in-depth understanding of extreme-scale I/O behavior and the I/O performance modeling and prediction are essential tools of the large-scale I/O evaluation process for addressing the needs of extreme-scale hybrid workloads. In this survey article, we focus on the state-of-the-art of the I/O behavior and performance analysis process for HPC systems in a 5-year time window and identify future research challenges.

Neuwirth, Sarah↗
DOE OSTI2025

Direct Air Reactive Capture and Conversion for Utility-Scale Energy Storage (Final Report)

This final report for FEW0277 summarizes the work performed over the project performance period of October 2021 – March 2025. This project was funded under the “Reactive Capture and Conversion R&D” lab call released in FY2021. The goal of the project was to develop dual-function materials and process for capturing CO 2 from the atmosphere and converting it into CH 4 . The work was organized into four parallel tracks in 1) direct air capture materials synthesis and characterization, 2) catalysts for CO 2 conversion, 3) mechanistic investigations via ab initio simulations, and 4) process modeling, technoeconomic analysis, and lifecycle assessment. The project was split into two budget periods. The first budget period focused on development of amine-based materials, due to their known performance for CO 2 direct air capture and their potential to act synergistically with metal catalysts to enable a low-temperature methanation pathway. The second budget period focused on development of alkali-based materials and a simulated-moving-bed process for high conversion catalytic reduction of captured CO 2 to CH 4 . All project milestones were completed during the project performance period and are summarized in this report. Our work resulted in publication of eight peer-reviewed manuscripts, one patent application, and numerous presentations given at domestic and international conferences and invited academic department seminars.

03 NATURAL GAS↗
DOE OSTI2026

Neoclassical transport and profile prediction in transport barriers

Strong gradient regions in tokamaks, such as the pedestal or internal transport barriers, are regions of reduced turbulence where neoclassical transport can play a dominant role. However, standard neoclassical transport theory assumes that the gradient length scales of density, temperature, and potential are of the order of the system size. In the pedestal, gradient length scales are much shorter and are measured to be of the order of the ion poloidal gyroradius. We present an extension of neoclassical theory that is applicable in transport barriers of large aspect ratio tokamaks. We show that particle and momentum transport are connected in such a way that a source of parallel momentum can drive a significant neoclassical ion particle flux. In strong gradient regions, density, electric potential, mean parallel flow, and ion temperature are shown to no longer be flux functions. Instead, they have a small but important poloidally varying piece that modifies the transport equations to lowest order. This introduces a nonlinearity in the transport problem through the coupling with quasineutrality that yields multiple co-existing solutions when solving for the plasma profiles. The different solutions could be connected to low and high transport states and jumps between solutions could provide a new neoclassical explanation for H-L back-transitions.

Trinczek, Silvia [Princeton Plasma Physics Laborat↗
DOE OSTI2019

Strain-induced spontaneous Hall effect in an epitaxial thin film of a Luttinger semimetal

Pyrochlore iridates have offered a plethora of novel phenomena owing to the combination of topology and correlation. Among them, much attention has been paid to P r 2 I r 2 O 7 , as it is known as a Luttinger semimetal characterized by quadratic band touching at the Brillouin zone center, suggesting that the topology of its electronic states can be tuned by a moderate lattice strain and external magnetic field. In this work, we report that our epitaxial P r 2 I r 2 O 7 thin films grown by solid-state epitaxy exhibit a spontaneous Hall effect that persists up to 50 K without having spontaneous magnetization within our experimental accuracy. This indicates that the system breaks the time reversal symmetry at a temperature scale that is too high for the magnetism to be due to Pr 4f moments and must be related to magnetic order of the iridium 5d electrons. Moreover, our analysis finds that the chiral anomaly induces the negative contribution to the magnetoresistance only when a magnetic field and the electric current are parallel to each other. Our findings suggest that the strained part of the thin film forms a magnetic Weyl semimetal state.

36 MATERIALS SCIENCE↗
DOE OSTI2024

MultiGreen: A multiplexing architecture for GreenGate cloning

Genetic modification of plants fundamentally relies upon customized vector designs. The ever-increasing complexity of transgenic constructs has led to increased adoption of modular cloning systems for their ease of use, cost effectiveness, and rapid prototyping. GreenGate is a modular cloning system catered specifically to designing bespoke, single transcriptional unit vectors for plant transformation—which is also its greatest flaw. MultiGreen seeks to address GreenGate’s limitations while maintaining the syntax of the original GreenGate kit. The primary limitations MultiGreen addresses are 1) multiplexing in series, 2) multiplexing in parallel, and 3) repeated cycling of transcriptional unit assembly through binary intermediates. MultiGreen efficiently concatenates bespoke transcriptional units using an additional suite of level 1acceptor vectors which serve as an assembly point for individual transcriptional units prior to final, level 2, condensation of multiple transcriptional units. Assembly with MultiGreen level 1 vectors scales at a maximal rate of 2*⌈ log 6 n ⌉+3 days per assembly, where n represents the number of transcriptional units. Further, MultiGreen level 1 acceptor vectors are binary vectors and can be used directly for plant transformation to further maximize prototyping speed. MultiGreen is a 1:1 expansion of the original GreenGate architecture’s grammar and has been demonstrated to efficiently assemble plasmids with multiple transcriptional units. MultiGreen has been validated by using a truncated violacein operon from Chromobacterium violaceum in bacteria and by deconstructing the RUBY reporter for in planta functional validation. MultiGreen currently supports many of our in-house multi transcriptional unit assemblies and will be a valuable strategy for more complex cloning projects.

Science & Technology - Other Topics↗
DOE OSTI2025

Data for Kim et al., "Variations in the optical and molecular composition of dissolved organic matter exported from coastal wetlands"

Knowledge about sources and composition of marsh-derived dissolved organic matter (DOM) is critical for understanding the role of marshes in coastal biogeochemical cycling and the fate of marsh-derived DOM in the ocean. To investigate tidal variability in composition of marsh-derived DOM, Kim et al. examined the optical and molecular characteristics of hourly surface water samples at three tidal creeks in the Chesapeake Bay. Groundwater samples along the terrestrial landscape gradient as well as estuarine water from the adjacent estuary at each site were also collected to help resolve sources of surface water DOM. Samples were collected in summer 2024 at three sites – SWH: Sweet Hall Marsh, GCW: Kirkpatrick Marsh, and GWI: Goodwin Islands – which are part of synoptic sites in the Chesapeake Bay region of the COMPASS-FME (Coastal Observations, Mechanisms, and Predictions Across Systems and Scales - Field, Measurements, and Experiments) project. Surface water samples were collected hourly over a 48-hour period at each site. Groundwater and estuarine water samples were collected once. This dataset includes- Surface water depth and salinity- Dissolved organic carbon (DOC) and total dissolved nitrogen (TDN) concentrations- Optical indices and relative composition of parallel factor analysis (PARAFAC) components- High resolution mass spectrometry data.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2017

Block-Parallel Data Analysis with DIY2

DIY2 is a programming model and runtime for block-parallel analytics on distributed-memory machines. Its main abstraction is block-structured data parallelism: data are decomposed into blocks; blocks are assigned to processing elements (processes or threads); computation is described as iterations over these blocks, and communication between blocks is defined by reusable patterns. By expressing computation in this general form, the DIY2 runtime is free to optimize the movement of blocks between slow and fast memories (disk and flash vs. DRAM) and to concurrently execute blocks residing in memory with multiple threads. This enables the same program to execute in-core, out-of-core, serial, parallel, single-threaded, multithreaded, or combinations thereof. This paper describes the implementation of the main features of the DIY2 programming model and optimizations to improve performance. DIY2 is evaluated on benchmark test cases to establish baseline performance for several common patterns and on larger complete analysis codes running on large-scale HPC machines.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Parallel Diffusion Coefficient of Energetic Charged Particles in the Inner Heliosphere from the Turbulent Magnetic Fields Measured by Parker Solar Probe

Diffusion coefficients of energetic charged particles in turbulent magnetic fields are a fundamental aspect of diffusive transport theory but remain incompletely understood. In this work, we use quasi-linear theory to evaluate the spatial variation of the parallel diffusion coefficient κ ∥ from the measured magnetic turbulence power spectra in the inner heliosphere. We consider the magnetic field and plasma velocity measurements from Parker Solar Probe made during Orbits 5–13. The parallel diffusion coefficient is calculated as a function of radial distance from 0.062 to 0.8 au, and the particle energy from 100 keV to 1 GeV. We find that κ ∥ increases exponentially with both heliocentric distance and energy of particles. The fluctuations in κ ∥ are related to the episodes of large-scale magnetic structures in the solar wind. By fitting the results, we also provide an empirical formula of κ ∥ = (5.16 ± 1.22) × 10 18 r 1.17 ± 0.08 E 0.71 ± 0.02 (cm 2 s -1 ) in the inner heliosphere, which can be used as a reference in studying the transport and acceleration of solar energetic particles as well as the modulation of cosmic rays.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Scalable PETSc Implementation of the Three-Phase Unbalanced AC Power Flow Solver

A revolutionary transition of energy infrastructure, spanning across supply and demand sectors, is happening all around the world. Integration of renewable energy sources into our ever-complex power distribution networks is central to the infrastructure planning of the future. One step towards achieving this goal is to solve unbalanced transmission and distribution systems. Traditional approaches and software tools are not suitable because they assume a completely balanced 3-phase system, hence a single-phase equivalent can be solved. Renewable energy such as wind and solar power are intermittent and can pose a challenge when integrating into an already unbalanced power grid. Furthermore, these power grids can be quite large at scale consisting of potentially thousands of buses, which can be computationally demanding. This poster presents preliminary work on designing a robust and scalable 3-phase unbalanced AC power flow solver. The proposed power flow solver will eventually be built on top of the PETSc library and utilizes state-of-the-art parallel data structures like the DMNetwork as well as iterative solvers and preconditioners.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2024

Accelerating Application Bulk Synchronous Writes in HPC Environments

High-bandwidth storage tiers are becoming more common for their capability to absorb high-rate, bursty I/Os. Notably, the designs of these fast storage tiers differ from system to system. The variation of these layers and non-uniform methods of access can pose chal- lenges for applications seeking to run at multiple HPC facilities. Therefore, in this work, we present Spectral, a rapid-output ab- straction library to accelerate application, bulk-synchronous writes on HPC systems. We design Spectral to enable applications to use high-bandwidth storage, such as node-local storage and dis- tributed, write-caches (e.g., burst buffers) transparently without requiring modifications to the application or file system source code. The key idea is to allow applications to spend most of the time performing productive work and to not require any source code changes for maximum portability on different HPC archi- tectures. Spectral internally re-routes write-only files through available, high-performance I/O resources before ultimately mi- grating them to the shared global parallel file system. For instance, on Summit, Spectral transparently places application outputs on node-local storage and then utilizes asynchronous migration to the center-wide GPFS file system. We evaluate Spectral on the Summit HPC system (1024 nodes) using the IOR benchmark and real scientific applications. Spectral shows linear performance scaling, improving application write performance by over an order of magnitude when compared to GPFS.

Khan, Awais↗