Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed parallelization”

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 1,585 records · Page 88

An application-level parallel I/O library for Earth system models

We describe the design and implementation of an application-level parallel I/O (PIO) library for the reading and writing of distributed arrays to several common scientific data formats. PIO provides the flexibility to control the number of I/O tasks through data rearrangement to an I/O friendly decomposition. This flexibility enables reductions in per task memory usage and improvements in disk I/O performance versus a serial I/O approach. We illustrate the impact various features within PIO have on memory usage and disk I/O bandwidth on a Cray XT5 system.

Computer Science↗

Direct numerical simulations for hybrid rocket boundary layers: Performance modeling and scaling

This paper presents a comprehensive performance and scaling analysis of direct numerical simulations for reacting boundary layers, focusing on slab burner configurations. Using a PETSc-based finite volume CFD framework, the study evaluates the scalability and computational cost of flow, chemistry, and radiation evaluations across 2D and 3D simulations. Polymethyl methacrylate (PMMA) is the fuel with pure O 2 as the oxidizer, modeled using a detailed chemical kinetics mechanism with 113 species and 660 reactions. A ray-tracing-based radiation solver, designed for distributed memory applications, is implemented to model radiation heat transfer. Parallel scalability is analyzed for the coupled flow, chemistry, and radiation heat transfer processes. Weak and strong scaling studies are conducted on up to 15,000 computational ranks, revealing robust performance when flow cells exceed 200 per rank. Chemistry evaluations dominate the computational cost in large 3D simulations, accounting for approximately 40% of the total runtime, while flow processes contribute around 35%, and radiation solver contributions remain below 10% due to reduced evaluation frequencies. GPU accelerated chemistry evaluation, implemented with Zero-RK, demonstrates significant promise, achieving up to a 4x speedup for workloads exceeding 30,000 cells per GPU. However, diminishing returns are observed for smaller workloads due to CPU-GPU communication overhead. This study identifies key challenges, including memory bottlenecks and the effects of domain partitioning on flow scalability, while highlighting the potential of GPU-accelerated chemistry to reduce computational costs. In conclusion, these findings provide realizable run configurations for 2D, 3D, and GPU-accelerated cases, offering insights for optimizing reactive flow solvers.

CFD Scalability↗

EQSIM—A multidisciplinary framework for fault-to-structure earthquake simulations on exascale computers, part II: Regional simulations of building response

The existing observational database of the regional-scale distribution of strong ground motions and measured building response for major earthquakes continues to be quite sparse. As a result, details of the regional variability and spatial distribution of ground motions, and the corresponding distribution of risk to buildings and other infrastructure, are not comprehensively understood. Utilizing high-performance computing platforms, emerging high-resolution, physics-based ground motion simulations can now resolve frequencies of engineering interest and provide detailed synthetic ground motions at high spatial density. This provides an opportunity for new insight into the distribution of infrastructure seismic demands and risk. In the work presented herein, the EQSIM fault-to-structure computational framework described in a companion paper, McCallen et al., is employed to investigate the regional-scale response of buildings to large earthquakes. A representative M = 7.0 strike-slip event is used to explore the distribution and amplitude of building demand, and comparisons are made between building response computed with fault-to-structure simulations and building response computed with existing measured near-fault earthquake records. New information on the distribution and variability of building response from high-performance parallel simulations is described and analyzed, and favorable first comparisons between building response predicted with both fault-to-structure simulations and real ground motions records are presented.

58 GEOSCIENCES↗

Physics of Intrinsic Rotation in Flux-Driven ITG Turbulence

Global, heat flux-driven ITG gyrokinetic simulations which manifest the formation of macroscopic, mean toroidal flow profiles with peak thermal Mach number 0.05, are reported. Both a particle-in-cell (XGC1p) and a semi-Lagrangian (GYSELA) approach are utilized without a priori assumptions of scale-separation between turbulence and mean fields. Flux-driven ITG simulations with different edge flow boundary conditions show in both approaches the development of net unidirectional intrinsic rotation in the co-current direction. Intrinsic torque is shown to scale approximately linearly with the inverse scale length of the ion temperature gradient. External momentum input is shown to effectively cancel the intrinsic rotation profile, thus confirming the existence of a local residual stress and intrinsic torque. Fluctuation intensity, intrinsic torque and mean flow are demonstrated to develop inwards from the boundary. The measured correlations between residual stress and two fluctuation spectrum symmetry breakers, namely E x B shear and intensity gradient, are similar. Avalanches of (positive) heat flux, which propagate either outwards or inwards, are correlated with avalanches of (negative) parallel momentum flux, so that outward transport of heat and inward transport of parallel momentum are correlated and mediated by avalanches. The probability distribution functions of the outward heat flux and the inward momentum flux show strong structural similarity

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Optimizing Extender Code for NCSX Analyses

Extender is a parallel C++ code for calculating the magnetic field in the vacuum region of a stellarator. The code was optimized for speed and augmented with tools to maintain a specialized NetCDF database. Two parallel algorithms were examined. An even-block work-distribution scheme was comparable in performance to a master-slave scheme. Large speedup factors were achieved by representing the plasma surface with a spline rather than Fourier series. The accuracy of this representation and the resulting calculations relied on the density of the spline mesh. The Fortran 90 module db access was written to make it easy to store Extender output in a manageable database. New or updated data can be added to existing databases. A generalized PBS job script handles the generation of a database from scratch

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Thermal effects in face seals.

Incompressible fluid laminar and turbulent flow between face seal rotating parallel surfaces, considering conduction, convection and dissipation influence on thermal distribution

Sneck, H. J.↗

Topology of induced lunar magnetic fields

Using the asymmetric theory of lunar induction the total and induced magnetic field line structure within the Moon and the diamagnetic cavity were obtained. Total field distributions are shown for orientations of the oscillating interplanetary field parallel, perpendicular and at 45 deg to the cavity axis. Induced field lines are shown only for the orientations of the interplanetary field parallel and orthogonal to the cavity axis. When compared with the field lines derived using the long wavelength limit of spherically symmetric vacuum induction theory, the configurations obtained using the asymmetric theory exhibit significant distortion. For all orientations of the interplanetary field, the field lines are strongly compressed on the sunlit hemisphere because of the confining solar wind pressure at the lunar surface and the exclusion of the field by the lunar core.

Schwartz, K.↗

Molecular beam mass spectrometer development

An analytical model, based on the kinetics theory of a drifting Maxwellian gas is used to determine the nonequilibrium molecular density distribution within a hemispherical shell open aft with its axis parallel to its velocity. The concept of a molecular shield in terrestrial orbit above 200 km is also analyzed using the kinetic theory of a drifting Maxwellian gas. Data are presented for the components of the gas density within the shield due to the free stream atmosphere, outgassing from the shield and enclosed experiments, and atmospheric gas scattered off a shield orbiter system. A description is given of a FORTRAN program for computating the three dimensional transition flow regime past the space shuttle orbiter that employs the Monte Carlo simulation method to model real flow by some thousands of simulated molecules.

Brock, F. J.↗

Some correlation of rock exposure ages and regolith dynamics

Exposure age information on lunar rocks and regolith turnover rates are correlated. If plotted in a cumulative fashion, the distribution of spallogenic noble-gas exposure ages is remarkably parallel to the rate at which various fractions of the regolith surface are cratered and/or excavated. It appears that the rate at which lunar rocks are excavated from within the regolith is strongly controlled by the impact environment. Some suggestions for future refinement of regolith dynamics are presented.

Horz, F.↗

Electron heating within the earth's bow shock

High-temporal-resolution measurements of electron velocity distributions have been obtained for many transits through the earth's bow shock. Within all oblique shocks studied, the maximum of the electron velocity distribution is offset with respect to the ion rest frame parallel to the magnetic field vector and directed downstream. These observations indicate that electron thermalization within the bow shock consists first of a downstream acceleration parallel to the magnetic field vector by the macroscopic shock electric field, followed by beam-driven plasma instabilities.

Feldman, W. C.↗

Interplanetary magnetic field effects on high latitude ionospheric convection

Relations between the electric field and the electric current in the ionosphere can be established on the basis of a system of mathematical and physical equations provided by the equations of current continuity and Ohm's law. For this reason, much of the synthesis of electric field and plasma velocity data in the F-region is made with the aid of similar data sets derived from field-aligned current and horizontal current measurements. During the past decade, the development of a self-consistent picture of the distribution and behavior of these measurements has proceeded almost in parallel. The present paper is concerned with the picture as it applies to the electric field and plasma drift velocity and its dependence on the interplanetary magnetic field. Attention is given to the southward interplanetary magnetic field and the northward interplanetary magnetic field.

Heelis, R. A.↗

Ion specific differences in energetic field aligned upflowing ions at 1 earth radius

An ion-upflow event over the northern auroral zone during a small magnetic storm on August 24, 1976 is characterized on the basis of S3-3 satellite ion-mass-spectrometer observations at 0.5-16 keV. The data are presented in tables, graphs, and diagrams and analyzed in detail. The distributions of upward-flowing O(+) and H(+) associated with parallel electric fields below the satellite are shown to have transverse temperatures of about 1 keV and 0.1 keV, respectively, while the corresponding parallel temperatures were about 1 keV and 0.6 keV; the average energy of the O(+) ions is found to be 2-3 times greater than that of the H(+) ions in the acceleration regions. Possible interpretations of these findings are discussed.

Ghielmetti, A. G.↗

Cosmic-ray hydrodynamics at relativistic shocks

A test particle description of first-order Fermi acceleration of cosmic rays at plane-parallel relativistic shocks is presented in which the pitch angle distribution is obtained explicitly and used to determine the pressure anisotropy or variable Eddington factor as a function of position. The Eddington factor is in turn used to find the hydrodynamical cosmic-ray sound speed, taking the anisotropy into account. For upstream flows speeds above c/sq rt 3, it is shown that the cosmic-ray sound speed exceeds the shock speed in a region just upstream of the shock and decreases montonically with distance further upstream, reaching the value c/sq rt 3 far from the shock. Thus, the shock succeeds in maintaining an anisotropy in the cosmic-ray distribution which is sufficient to keep the flow subsonic with respect to the cosmic-ray sound speed both downstream and in a finite region upstream. The astrophysical implications of these findings are discussed.

Kirk, J. G.↗

Parallel arrays of Josephson junctions for submillimeter local oscillators

In this paper we discuss the influence of the DC biasing circuit on operation of parallel biased quasioptical Josephson junction oscillator arrays. Because of nonuniform distribution of the DC biasing current along the length of the bias lines, there is a nonuniform distribution of magnetic flux in superconducting loops connecting every two junctions of the array. These DC self-field effects determine the state of the array. We present analysis and time-domain numerical simulations of these states for four biasing configurations. We find conditions for the in-phase states with maximum power output. We compare arrays with small and large inductances and determine the low inductance limit for nearly-in-phase array operation. We show how arrays can be steered in H-plane using the externally applied DC magnetic field.

Pance, Aleksandar↗

Supercomputing '91; Proceedings of the 4th Annual Conference on High Performance Computing, Albuquerque, NM, Nov. 18-22, 1991

Various papers on supercomputing are presented. The general topics addressed include: program analysis/data dependence, memory access, distributed memory code generation, numerical algorithms, supercomputer benchmarks, latency tolerance, parallel programming, applications, processor design, networks, performance tools, mapping and scheduling, characterization affecting performance, parallelism packaging, computing climate change, combinatorial algorithms, hardware and software performance issues, system issues. (No individual items are abstracted in this volume)

Source record↗

Boundary element analysis on vector and parallel computers

Boundary element analysis (BEA) can be characterized as a numerical technique that generally shifts the computational burden in the analysis toward numerical integration and the solution of nonsymmetric and either dense or blocked sparse systems of algebraic equations. Researchers have explored the concept that the fundamental characteristics of BEA can be exploited to generate effective implementations on vector and parallel computers. In this paper, the results of some of these investigations are discussed. The performance of overall algorithms for BEA on vector supercomputers, massively data parallel single instruction multiple data (SIMD), and relatively fine grained distributed memory multiple instruction multiple data (MIMD) computer systems is described. Some general trends and conclusions are discussed, along with indications of future developments that may prove fruitful in this regard.

Kane, J. H.↗

Parallelization of Program to Optimize Simulated Trajectories (POST3D)

This paper describes the parallelization of the Program to Optimize Simulated Trajectories (POST3D). POST3D uses a gradient-based optimization algorithm that reaches an optimum design point by moving from one design point to the next. The gradient calculations required to complete the optimization process, dominate the computational time and have been parallelized using a Single Program Multiple Data (SPMD) on a distributed memory NUMA (non-uniform memory access) architecture. The Origin2000 was used for the tests presented.

Hammond, Dana P.↗

Benchmarking Memory Performance with the Data Cube Operator

Data movement across a computer memory hierarchy and across computational grids is known to be a limiting factor for applications processing large data sets. We use the Data Cube Operator on an Arithmetic Data Set, called ADC, to benchmark capabilities of computers and of computational grids to handle large distributed data sets. We present a prototype implementation of a parallel algorithm for computation of the operatol: The algorithm follows a known approach for computing views from the smallest parent. The ADC stresses all levels of grid memory and storage by producing some of 2d views of an Arithmetic Data Set of d-tuples described by a small number of integers. We control data intensity of the ADC by selecting the tuple parameters, the sizes of the views, and the number of realized views. Benchmarking results of memory performance of a number of computer architectures and of a small computational grid are presented.

Frumkin, Michael A.↗