Search NASA⌕ Search

SEARCH · Search NASA

Results for “bundle method”

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 343 records · Page 19

Future Opportunities for LWR Irradiations in US Test Reactors

After several years of relatively low activity in the field of Light Water Reactor (LWR) fuel development, the Department of Energy again began to engage in developing new fuel technologies and irradiation performance data prompted by the Fukushima Daichi nuclear accidents. New competencies for irradiation testing in material test reactors in the United States began to be developed at this time using the Advanced Test Reactor (ATR), High Flux Isotope Reactor (HFIR), Massachusetts Institute of Technology Reactor (MITR), and the Transient Reactor Test Facility (TREAT). Capsules for testing fuel and cladding materials in ATR and HFIR were deployed, a Pressurized Water Reactor (PWR) condition loop for testing fuel rods was established in ATR, cladding corrosion studies were performed using a water loop in MITR, and TREAT pulse testing capabilities were commissioned for fuel rods in water capsules. The more recent and unexpected closure of the Halden Boiling Water Reactor (HBWR) also prompted further investments in Loss of Coolant Accident (LOCA) testing capabilities at TREAT. New configurations of these test devices show further potential in enhanced steam condition control and other investigations are building toward a flowing water loop for testing transient to dryout conditions. The closure of HBWR also prompted a major project currently underway to construct additional water loops in ATR where a novel approach is being pursued to enable Boiling Water Reactor (BWR) conditions. A meaningful collaborative project was awarded to MITR which, amidst an unexpected major overhaul of the reactor, has expanded cladding corrosion test capabilities at MITR. New explorations have led to methods for unique experiments at HFIR including channel box irradiations. New device developments are also bridging toward future potential for instrumented capsule irradiation tests in ATR and HFIR. Finally, a new project referred to as the System Physics Advanced Reactor Critical facility (SPARC) is gaining traction towards a large zero-power reactor able to produce physics validation data for LWR fuel bundle designs with increased enrichment and enhanced absorbers for 24-month operation cycles. This paper provides a brief summary of the status of these irradiation testbed capabilities with an emphasis on current efforts toward future capabilities to obtain new data and maximize the performance potential of LWR fuel technologies.

Woolstenhulme, Nicolas [Idaho National Laboratory ↗

Heat Emission and Temperature Non-uniformity in Rod Bundles with Different Spacing, Cooled by Heavy Liquid-Metal Coolant

Heat emission and temperature non-uniformity along the perimeter of rods in free lattices of fuel elements with heavy liquid-metal coolant are studied. The experimental data for an assembly of smooth rods, assemblies of fuel elements with spiral ribbing, and transverse lattices are analyzed. The studies show that there is almost no temperature non-uniformity in a lattice of smooth fuel elements along its perimeter, which is in contrast to the high general azimuthal temperature non-uniformity in an assembly with spacing by spiral coils. Rib spacing resulted in significant reduction of heat emission. In an assembly with transverse spacer lattices, the heat emission increases only in the region containing the lattices; in the spaces between lattices, it is approximately equal to the heat emission in a bundle with smooth fuel elements. Inside a spacer lattice, periodic temperature non-uniformities appear along the perimeter of the fuel elements; these non-uniformities are due to the fuel elements touching the elements of the lattice. A relation is recommended for calculating heat emission (Nusselt numbers) and temperature non-uniformity along the perimeter of the fuel elements for the studied methods of spacing the fuel elements.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Enamel nanocrystal misorientation increased with meat-eating and agriculture

Enamel covers teeth, is the hardest tissue in the vertebrate body and has a complex multiscale structure from nanometres to millimetres. The structure comprises thin, long hydroxyapatite (Ca 5 (PO 4 ) 3 OH) nanocrystals, 50–70 nm wide, many micrometres long, parallel and bundled into approximately 5-µm-wide rods. The rods undulate and cross into a microscale ‘decussation pattern’ that toughens enamel by deflecting cracks. However, the crystallographic orientation of enamel nanocrystals is poorly understood. Here we show that the misorientation angle of adjacent nanocrystals varies markedly across 12 primate teeth spanning 9 species, 17.8 million years of evolution and diverse diets. Using a method called Polarization Enabled Large Input of Crystal Angles at the Nanoscale (PELICAN), we compare nanocrystals in the same (pre)molar locations and show that misorientation increases with food hardness in extant and fossil non-human apes and monkeys. We compare misorientation across three major dietary shifts in human evolution: the transition to meat-eating about 2.0–1.5 million years before present, to agriculture (about 12,000 years before present), and the Industrial Revolution (about 250 years before present). We show that over the past 1.6 million years, in the human lineage misorientation increased with time, especially when meat and stone-ground grains were introduced into human diets, but not with the Industrial Revolution. Thus, besides macro-changes, teeth adapted to dietary change at the nanoscale and crystallographically. This observation suggests that misorientation may contribute to enamel’s resilience; thus, bioinspired materials may consider small misorientation angles for added resilience.

biomaterials↗

Subcellular-level resolution MALDI-MS imaging of maize leaf metabolites by MALDI-linear ion trap-Orbitrap mass spectrometer

A significant limiting factor in achieving high spatial resolution for matrix-assisted laser desorption ionization-mass spectrometry (MALDI-MS) imaging is the size of the laser spot at the sample surface. We present modifications to the beam-delivery optics of a commercial MALDI-linear ion trap-Orbitrap instrument, incorporating an external Nd:YAG laser, beam-shaping optics, and an aspheric focusing lens, to reduce the minimum laser spot size from ~50 μm for the commercial configuration down to ~9 μm for the modified configuration. This improved system was applied for MALDI-MS imaging of cross sections of juvenile maize leaves at 5-μm spatial resolution using an oversampling method. There are a variety of different metabolites including amino acids, glycerolipids, and defense-related compounds were imaged at a spatial resolution well below the size of a single cell. Such images provide unprecedented insights into the metabolism associated with the different tissue types of the maize leaf, which is known to asymmetrically distribute the reactions of C4 photosynthesis among the mesophyll and bundle sheath cell types. The metabolite ion images correlate with the optical images that reveal the structures of the different tissues, and previously known and newly revealed asymmetric metabolic features are observed.

47 OTHER INSTRUMENTATION↗

Effect of vortex hotspots on the radio-frequency surface resistance of superconductors

We present detailed experimental and theoretical investigations of hotspots produced by trapped vortex bundles and their effect on the radio-frequency (rf) surface resistance R s of superconductors at low temperatures. Our measurements of R s , combined with the temperature mapping and laser scanning of a 2.36-mm-thick Nb plate incorporated into a 3.3-GHz Nb resonator cavity cooled by the superfluid He at 2 K, revealed spatial scales and temperature distributions of hotspots and showed that they can be moved or split by thermal gradients produced by the scanning laser beam. These results, along with the observed hysteretic field dependence of R s which can be tuned by the scanning laser beam, show that the hotspots in our Nb sample are due to trapped vortex bundles which contain ~10 6 vortices spread over regions ~0.1–1 cm. We calculated the frequency dependence of the rf power dissipated by oscillating vortex segments trapped between nanoscale pinning centers, taking into account all bending modes and the nonlocal line tension of the vortex driven by rf Meissner currents. We also calculated the temperature distributions caused by trapped vortex hotspots, and suggested a method of reconstructing the spatial distribution of vortex dissipation sources from the observed temperature maps. Vortex hotspots can dominate the residual surface resistance at low temperatures and give rise to a significant dependence of R s on the rf field amplitude H p , which can have important implications for the rf resonating cavities used in particle accelerators and for thin-film structures used in quantum computing and photon detectors.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Fluid structure interaction modelling for the vibration of tube bundles, part I: analysis of the fluid flow in a tube bundle

It is well known that a fluid may strongly influence the dynamic behaviour of a structure. Many different physical phenomena may take place, depending on the conditions: fluid flow, fluid at rest, little or high displacements of the structure. Inertial effects can take place, with lower vibration frequencies, dissipative effects also, with damping, instabilities due to the fluid flow (Fluid Induced Vibration). In this last case the structure is excited by the fluid. Tube bundles structures are very common in the nuclear industry. The reactor cores and the steam generators are both structures immersed in a fluid which may be submitted to a seismic excitation or an impact. In this case the structure moves under an external excitation, and the movement is influence by the fluid. The main point in such system is that the geometry is complex, and could lead to very huge sizes for a numerical analysis. Homogenization models have been developed based on the Euler equations for the fluid. Only inertial effects are taken into account. A next step in the modelling is to build models based on the homogenization of the Navier-Stokes equations. The papers presents results on an important step in the development of such model: the analysis of the fluid flow in a oscillating tube bundle. The analysis are made from the results of simulations based on the Navier-Stokes equations for the fluid. Comparisons are made with the case of the oscillations of a single tube, for which a lot of results are available in the literature. Different fluid flow pattern may be found, depending in the Reynolds number (related to the velocity of the bundle) and the Keulegan Carpenter number (related to the displacement of the bundle). A special attention is paid to the quantification of the inertial and dissipative effects, and to the forces exchanges between the bundle and the fluid. The results of such analysis will be used in the building of models based on the homogenization of the Navier-Stokes equations for the fluid. (authors)

42 ENGINEERING↗

Electrostatic Contributions Drive the Interaction Between Staphylococcus aureus Protein Efb-C and its Complement Target C3d

The C3-inhibitory domain of Staphylococcus aureus extracellular fibrinogen-binding protein (Efb-C) defines a novel three-helix bundle motif that regulates complement activation. Previous crystallographic studies of Efb-C bound to its cognate subdomain of human C3 (C3d) identified Arg-131 and Asn-138 of Efb-C as key residues for its activity. In order to characterize more completely the physical and chemical driving forces behind this important interaction, we employed in this study a combination of structural, biophysical, and computational methods to analyze the interaction of C3d with Efb-C and the single-point mutants R131A and N138A. Our results show that while these mutations do not drastically affect the structure of the Efb-C/C3d recognition complex, they have significant adverse effects on both the thermodynamic and kinetic profiles of the resulting complexes. We also characterized other key interactions along the Efb-C/C3d binding interface and found an intricate network of salt bridges and hydrogen bonds that anchor Efb-C to C3d, resulting in its potent complement inhibitory properties.

08 HYDROGEN↗

Origin of Pulsed Radio Emission from Magnetars

Extended periods of radio pulsations have been observed for six magnetars, displaying characteristics different from those of ordinary pulsars. In this Letter, we argue that radio emission is generated in a closed, twisted magnetic flux bundle originating near the magnetic pole and extending beyond 100 km from the magnetar. The electron–positron flow in the twisted bundle has to carry electric current and, at the same time, experiences a strong drag from the radiation field of the magnetar. This combination forces the plasma into a “radiatively locked” state with a sustained two-stream instability, generating radio emission. We demonstrate this mechanism using novel first-principles simulations that follow the plasma behavior by solving the relativistic Vlasov equation with the discontinuous Galerkin method. First, using one-dimensional simulations, we demonstrate how radiative drag induces the two-stream instability, sustaining turbulent electric fields. When extended to two dimensions, the system produces electromagnetic waves, including superluminal modes capable of escaping the magnetosphere. We measure their frequency and emitted power and incorporate the local simulation results into a global magnetospheric model. The model explains key features of the observed radio emission from magnetars: its appearance after an X-ray outburst, wide pulse profiles, luminosities ∼10 30 erg s −1 , and a broad range of frequencies extending up to ∼100 GHz.

Astronomical simulations↗

Evaluation Of Risk And Possible Mitigation Schemes For Previously Unidentified Hazards

This report presents the results of arc track testing conducted to determine if such a transfer of power to un-energized wires is possible and/or likely during an arcing event, and to evaluate an array of protection schemes that may significantly reduce the possibility of such a transfer. The results of these experiments may be useful for determining the level of protection necessary to guard against spurious voltage and current being applied to safety critical circuits. It was not the purpose of these experiments to determine the probability of the initiation of an arc track event only if an initiation did occur could it cause the undesired event: an inadvertent thruster firing. The primary wire insulation used in the Orbiter is aromatic polyimide, or Kapton , a construction known to arc track under certain conditions [3]. Previous Boeing testing has shown that arc tracks can initiate in aromatic polyimide insulated 28 volts direct current (VDC) power circuits using more realistic techniques such as chafing with an aluminum blade (simulating the corner of an avionics box or lip of a wire tray), or vibration of an aluminum plate against a wire bundle [4]. Therefore, an arc initiation technique was chosen that provided a reliable and consistent technique of starting the arc and not a realistic simulation of a scenario on the vehicle. Once an arc is initiated, the current, power and propagation characteristics of the arc depend on the power source, wire gauge and insulation type, circuit protection and series resistance rather than type of initiation. The initiation method employed for these tests was applying an oil and graphite mixture to the ends of a powered twisted pair wire. The flight configuration of the heater circuits, the fuel/oxider (or ox) wire, and the RCS jet solenoid were modeled in the test configuration so that the behavior of these components during an arcing event could be studied. To determine if coil activation would occur with various protection wire schemes, 145 tests were conducted using various fuel/ox wire alternatives (shielded and unshielded) and/or different combinations of polytetrafuloroethylene (PTFE), Mystik tape and convoluted wraps to prevent unwanted coil activation. Test results were evaluated along with other pertinent data and information to develop a mitigation strategy for an inadvertent RCS firing. The SSP evaluated civilian aircraft wiring failures to search for aging trends in assessing the wire-short hazard. Appendix 2 applies Weibull statistical methods to the same data with a similar purpose.

Linzey, William↗

Pressure Drop in Seven-Pin Wire-Wrapped Rod Bundle for the Sodium Cartridge Loop in Versatile Test Reactor

This work studies the hydrodynamics of the seven pin wire-wrapped rod bundle in the sodium cartridge loop for the Versatile Test Reactor (VTR) through scaled water experiments and computational fluid dynamics (CFD) simulations. Here, the scaling analysis is first performed to demonstrate the hydrodynamic similarity between water and sodium flows at the same Reynolds number Re. A separate-effects test facility is designed and constructed based on the scaling analysis. Detailed experimental data on the pressure drop covering a wide range of Re values (1165 to 27 689) are obtained, which are used to evaluate existing correlations for friction factor and to benchmark CFD simulations. The experimentally determined friction factors f Exp agree well with the Upgraded Cheng and Todreas Detailed Correlation and Pacio-Chen-Todreas Detailed Model within ±7% but are significantly underpredicted by Rehme's correlation by 25%. Various CFD near-wall treatment methods are tested using ANSYS Fluent and evaluated by experimental data. It is found that when the recommended wall y + values are met, most of the near-wall treatment methods can give accurate friction factor predictions. The resolved near-wall method (y + ~ 1) with the Shear Stress Transport k - ω turbulence model and the scalable wall functions (y + > 12) with the realizable k - ε turbulence model can predict f Exp within ±5% The standard wall functions (y + > 30) and nonequilibrium wall functions (y + > 30) with the realizable k - ε model can predict f Exp within ±10%.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Extended Hamiltonian systems in multisymplectic field theories

We consider Hamiltonian systems in first-order multisymplectic field theories. We review the properties of Hamiltonian systems in the so-called restricted multimomentum bundle, including the variational principle which leads to the Hamiltonian field equations. In an analogous way to how these systems are defined in the so-called extended (symplectic) formulation of nonautonomous mechanics, we introduce Hamiltonian systems in the extended multimomentum bundle. The geometric properties of these systems are studied, the Hamiltonian equations are analyzed using integrable multivector fields, the corresponding variational principle is also stated, and the relation between the extended and the restricted Hamiltonian systems is established. All these properties are also adapted to certain kinds of submanifolds of the multimomentum bundles in order to cover the case of almost-regular field theories.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Effect of tetrahedral amorphous carbon coating on the resistivity and wear of single-walled carbon nanotube network

Single walled carbon nanotube networks (SWCNTNs) were coated by tetrahedral amorphous carbon (ta-C) to improve the mechanical wear properties of the composite film. The ta-C deposition was performed by using pulsed filtered cathodic vacuum arc method resulting in the generation of C+ ions in the energy range of 40–60 eV which coalesce to form a ta-C film. The primary disadvantage of this process is a significant increase in the electrical resistance of the SWCNTN post coating. The increase in the SWCNTN resistance is attributed primarily to the intrinsic stress of the ta-C coating which affects the inter-bundle junction resistance between the SWCNTN bundles. E-beam evaporated carbon was deposited on the SWCNTNs prior to the ta-C deposition in order to protect the SWCNTN from the intrinsic stress of the ta-C film. The causes of changes in electrical resistance and the effect of evaporated carbon thickness on the changes in electrical resistance and mechanical wear properties have been studied.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Cosmic Ray Muon Radiography of Spent Fuel in Dry Storage Casks

In the absence of a permanent repository or reprocessing facilities, highly radioactive spent nuclear fuel bundles are typically placed in long-term storage in above ground casks. These dry storage casks are heavily shielded to prevent the escape of radiation from the contents to the environment. Casks under international safeguards are monitored to detect diversion of spent fuel. The heavy shielding of the cask walls also precludes inspection of cask contents with typical radiographic probes. When international safeguards inspectors need to verify cask contents (which may be necessary to recover from a loss of continuity of knowledge), the casks must be moved back into a cooling pool to be safely opened and visually inspected. This time-consuming procedure is invasive and potentially dangerous. A non-destructive radiographic method to verify a cask's contents in situ would therefore be a useful safeguards tool. Highly energetic cosmic ray muons are an alternative tomographic probe. Since muons are colorless leptons, they do not participate in strong interactions with nucleons, and their large mass suppresses radiative energy loss while passing through materials. These properties allow muons to penetrate dense shielding that defeats more typical probes like photons and neutrons. Measurements of the scattering of muons that pass through objects provide information on the object's internal structure. This summary describes the first measurements using cosmic ray muons to inspect the contents of a partially loaded dry storage cask. We show that partial defects in the cask loading can be identified with this technique. This data shows that cosmic ray muons are an effective radiographic probe of dry storage casks. This technique may be useful for international safeguards inspectors wishing to verify a cask's contents without opening it. The analysis techniques described here allow the column of missing fuel bundles to be determined, but not the row. Recording data from several positions around the cask would allow ambiguities in x-y position to be resolved. Alternatively, a detector could be designed such that it surrounds the cask and provides a full image of the casks interior. Improved measurements and the design of optimized cask imaging instruments are being explored. (authors)

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Automatic Calibration and Health Monitoring of Infrastructure Sensors

Smart transportation infrastructure relies on networks of heterogeneous sensors - cameras, radars, and lidars - continuously monitoring traffic conditions. However, executing the initial spatial calibration of multiple sensors and the subsequent health monitoring presents significant operational challenges. Environmental factors, mechanical vibrations, and gradual drift cause spatial misalignment, degrading fusion performance and tracking accuracy. Traditional calibration approaches require manual intervention with specialized targets or survey equipment, resulting in service interruptions and high maintenance costs. This work presents an automated framework for initial calibration and continuous health monitoring without human intervention or service disruption. Our approach addresses two critical problems: (1) detecting when sensors become miscalibrated during operation, and (2) automatically re-establishing spatial alignment using only operational traffic data. The health monitoring component analyzes measurement innovations - differences between sensor observations and predicted object states - to detect systematic biases indicative of calibration drift. By computing bias magnitude, directional consistency, and rejection rates, the system identifies miscalibrations as small as 0.5 meters. Unlike traditional methods requiring known calibration targets, our diagnostic operates continuously on live traffic observations, enabling early detection before fusion quality degrades. The automatic recalibration algorithm leverages overlapping sensor fields-of-view and temporal correlation of vehicle observations. Using graph-based optimization, the system automatically discovers which sensor pairs observe common regions, estimates pairwise spatial transformations using RANSAC-based robust estimation, and jointly optimizes all sensor poses through bundle adjustment. The framework handles practical deployment challenges, including different sensor sampling rates (1-10 Hz), varying installation positions, unknown orientations, and limited overlap regions (>10%). When approximate sensor positions are available from installation surveys (+/-1m accuracy), the algorithm additionally estimates sensor orientations, refining both position and rotation to sub-meter and sub-degree accuracy. We validate the framework on multi-hour traffic datasets from six heterogeneous sensors with sampling rates ranging from 1 Hz to 10 Hz. Results demonstrate successful calibration even with sparse overlap (<20%) and automatic detection of miscalibrations exceeding 0.8 meters. This work enables a "deploy-and-forget" sensor infrastructure that maintains calibration autonomously, reducing maintenance costs while improving tracking accuracy. The techniques generalize beyond transportation to any multi-sensor monitoring application requiring robust spatial alignment, including smart cities, industrial monitoring, and surveillance systems.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Universal corner symmetry and the orbit method for gravity

A universal symmetry algebra organizing the gravitational phase space has been recently found. It corresponds to the subset of diffeomorphisms that become physical at corners – codimension-2 surfaces supporting Noether charges. It applies to both finite distance and asymptotic corners. In this paper, we study this algebra and its representations, via the coadjoint orbit method. We show that generic orbits of the universal algebra split into sub-orbits spanned by finite distance and asymptotic corner symmetries, such that the full universal symmetry algebra gives rise to a unified treatment of corners in a manifold. We then identify the geometric structure that captures these algebraic properties on corners, which is the Atiyah Lie algebroid associated to a principal GL(2,R) $\ltimes$ R 2 -bundle. This structure is suggestive of the existence of a novel quantum gravitational theory which would unitarily glue such geometric structures, with spacetime geometries appearing as semi-classical configurations.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Development of the tailored SiC/SiC composites by the combined fabrication process of ICVI and NITE methods

In order to improve the thermo-mechanical performances of SiC/SiC composite, process improvement and modification by the combination of nano-infiltration and transient eutectic-phase (NITE) method and chemical vapor infiltration (CVI) method were studied. Multilayered PyC/SiC fiber coating and matrix infiltration within fiber-tows were prepared with isothermal/isobaric CVI (ICVI) method and full-densification of SiC matrix was examined with NITE methods using four kinds of processing options. Applied pressure was useful for nearly-full matrix densification due to the promoting infiltration driving force of SiC nano-powder intra-fiber-tows, but simultaneously caused the sever degradation of fibers and interphase with fracture, resulting in lower strength. Increase of additives amount and additional polymer were effective ways for matrix densification by SiC nano-power infiltration intra-fiber bundles without pressure. Thermal conductivity was greatly improved with the decrease of matrix porosity. The tailoring of thermo-mechanical properties might be easily controlled by the SiC matrix porosity without processinduced fibers and interphases degradations.

Shimoda, Kazuya↗

Using coronal loops to reconstruct the magnetic field of an active region before and after a major flare

The shapes of solar coronal loops are sensitive to the presence of electrical currents that are the carriers of the non-potential energy available for impulsive activity. We use this information in a new method for modeling the coronal magnetic field of active region (AR) 11158 as a nonlinear force-free field (NLFFF). The observations used are coronal images around the time of major flare activity on 2011 February 15, together with the surface line-of-sight magnetic field measurements. The data are from the Helioseismic and Magnetic Imager and Atmospheric Imaging Assembly on board the Solar Dynamics Observatory. The model fields are constrained to approximate the coronal loop configurations as closely as possible, while also being subject to the force-free constraints. The method does not use transverse photospheric magnetic field components as input and is thereby distinct from methods for modeling NLFFFs based on photospheric vector magnetograms. We validate the method using observations of AR 11158 at a time well before major flaring and subsequently review the field evolution just prior to and following an X2.2 flare and associated eruption. The models indicate that the energy released during the instability is about 1 × 10{sup 32} erg, consistent with what is needed to power such a large eruptive flare. Immediately prior to the eruption, the model field contains a compact sigmoid bundle of twisted flux that is not present in the post-eruption models, which is consistent with the observations. The core of that model structure is twisted by ≈0.9 full turns about its axis.

79 ASTRONOMY AND ASTROPHYSICS↗

Designing a graphics processing unit accelerated petaflop capable lattice Boltzmann solver: Read aligned data layouts and asynchronous communication

The lattice Boltzmann method is a well-established numerical approach for complex fluid flow simulations. Recently, general-purpose graphics processing units (GPUs) have become available as high-performance computing resources at large scale. We report on designing and implementing a lattice Boltzmann solver for multi-GPU systems that achieves 1.79 PFLOPS performance on 16,384 GPUs. To achieve this performance, we introduce a GPU compatible version of the so-called bundle data layout and eliminate the halo sites in order to improve data access alignment. Furthermore, we make use of the possibility to overlap data transfer between the host central processing unit and the device GPU with computing on the GPU. As a benchmark case, we simulate flow in porous media and measure both strong and weak scaling performance with the emphasis being on large-scale simulations using realistic input data.

Computer Science↗