Search NASA⌕ Search

SEARCH · Search NASA

Results for “partitioned methods”

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 559 records · Page 31

Reduce Operations: Send Volume Balancing While Minimizing Latency

Communication hypergraph model was introduced in a two-phase setting for encapsulating multiple communication cost metrics (bandwidth and latency), which are proven to be important in parallelizing irregular applications. In the first phase, computational-task-to-processor assignment is performed with the objective of minimizing total volume while maintaining computational load balance. In the second phase, communication-task-to-processor assignment is performed with the objective of minimizing total number of messages while maintaining communication-volume balance. The reduce-communication hypergraph model suffers from failing to correctly encapsulate send-volume balancing. We propose a novel vertex weighting scheme that enables part weights to correctly encode send-volume loads of processors for send-volume balancing. The model also suffers from increasing the total communication volume during partitioning. To decrease this increase, we propose a method that utilizes the recursive bipartitioning framework and refines each bipartition by vertex swaps. For performance evaluation, we consider column-parallel SpMV, which is one of the most widely known applications in which the reduce-task assignment problem arises. Extensive experiments on 313 matrices show that, compared to the existing model, the proposed models achieve considerable improvements in all communication cost metrics. Furthermore, these improvements lead to an average decrease of 30% in parallel SpMV time on 512 processors for 70 matrices with high irregularity.

97 MATHEMATICS AND COMPUTING↗

Development of parallel algorithms for electrical power management in space applications

The application of parallel techniques for electrical power system analysis is discussed. The Newton-Raphson method of load flow analysis was used along with the decomposition-coordination technique to perform load flow analysis. The decomposition-coordination technique enables tasks to be performed in parallel by partitioning the electrical power system into independent local problems. Each independent local problem represents a portion of the total electrical power system on which a loan flow analysis can be performed. The load flow analysis is performed on these partitioned elements by using the Newton-Raphson load flow method. These independent local problems will produce results for voltage and power which can then be passed to the coordinator portion of the solution procedure. The coordinator problem uses the results of the local problems to determine if any correction is needed on the local problems. The coordinator problem is also solved by an iterative method much like the local problem. The iterative method for the coordination problem will also be the Newton-Raphson method. Therefore, each iteration at the coordination level will result in new values for the local problems. The local problems will have to be solved again along with the coordinator problem until some convergence conditions are met.

Berry, Frederick C.↗

Air Traffic Sector Configuration Change Frequency

Several techniques for partitioning airspace have been developed in the literature. The question of whether a region of airspace created by such methods can be used with other days of traffic, and the number of times a different partition is needed during the day is examined in this paper. Both these aspects are examined for the Fort Worth Center airspace sectors. A Mixed Integer Linear Programming method is used with actual air traffic data of ten high-volume low-weather-delay days for creating sectors. Nine solutions were obtained for each two-hour period of the day by partitioning the center airspace into two through 18 sectors in steps of two sectors. Actual track-data were played back with the generated partitions for creating histograms of the traffic-counts. The best partition for each two-hour period was then identified based on the nine traffic-count distributions. Numbers of sectors in such partitions were analyzed to determine the number of times a different configuration is needed during the day. One to three partitions were selected for the 24-hour period, and traffic data from ten days were played back to test if the traffic-counts stayed below the threshold values associated with these partitions. Results show that these partitions are robust and can be used for longer durations than they were designed for

Chatterji, Gano Broto↗

Additive Polynomial Time Integrators, Part I: Framework and Fully Implicit-Explicit Collocation Methods

In this paper we generalize the polynomial time integration framework to additively partitioned initial value problems. The framework we present is general and enables the construction of many new families of additive integrators with arbitrary order-of-accuracy and varying degree of implicitness. In this first work, we focus on a new class of implicit-explicit polynomial block methods that are based on fully implicit Runge–Kutta methods with Radau nodes and that possess high stage order. Here, we show that the new fully implicit-explicit (FIMEX) integrators have improved stability compared to existing IMEX Runge–Kutta methods, while also being more computationally efficient due to recent developments in preconditioning techniques for solving the associated systems of nonlinear equations. For PDEs on periodic domains where the implicit component is trivial to invert, we will show how parallelization of the right-hand side evaluations can be exploited to obtain significant speedup compared to existing serial IMEX Runge–Kutta methods. For parallel (in space) finite element discretizations, the new methods can achieve orders of magnitude better accuracy than existing IMEX Runge–Kutta methods and/or achieve a given accuracy several times times faster in terms of computational runtime.

97 MATHEMATICS AND COMPUTING↗

Point-in-convex polygon and point-in-convex polyhedron algorithms with O(1) complexity using space subdivision

There are many space subdivision and space partitioning techniques used in many algorithms to speed up computations. They mostly rely on orthogonal space subdivision, resp. using hierarchical data structures, e.g. BSP trees, quadtrees, octrees, kd-trees, bounding volume hierarchies etc. However in some applications a non-orthogonal space subdivision can offer new ways for actual speed up. In the case of convex polygon in E{sup 2} a simple Point-in-Polygon test is of the O(N) complexity and the optimal algorithm is of O(log N) computational complexity. In the E{sup 3} case, the complexity is O(N) even for the convex polyhedron as no ordering is defined. New Point-in-Convex Polygon and Point-in-Convex Polyhedron algorithms are presented based on space subdivision in the preprocessing stage resulting to O(1) run-time complexity. The presented approach is simple to implement. Due to the principle of duality, dual problems, e.g. line-convex polygon, line clipping, can be solved in a similarly.

97 MATHEMATICS AND COMPUTING↗

Approximate inverse-based block preconditioners in poroelasticity

We focus on the fully implicit solution of the linear systems arising from a three-field mixed finite element approximation of Biot’s poroleasticity equations. The objective is to develop algebraic block preconditioners for the efficient solution of such systems by Krylov subspace methods. In this work, we investigate the use of approximate inverse-based techniques to decouple the native system of equations and obtain explicit sparse approximations of the Schur complements related to the physics-based partitioning of the unknowns by field type. Here, the proposed methods are tested in various numerical experiments including real-world applications dealing with petroleum and geotechnical engineering.

54 ENVIRONMENTAL SCIENCES↗

Spectral evolution of two-dimensional kinetic plasma turbulence in the wavenumber-frequency domain

We present a method for studying the evolution of plasma turbulence by tracking dispersion relations in the energy spectrum in the wavenumber-frequency domain. We apply hybrid plasma simulations in a simplified two-dimensional geometry to demonstrate our method and its applicability to plasma turbulence in the ion kinetic regime. We identify four dispersion relations: ion-Bernstein waves, oblique whistler waves, oblique Alfvén/ion-cyclotron waves, and a zero-frequency mode. The energy partition and frequency broadening are evaluated for these modes. The method allows us to determine the evolution of decaying plasma turbulence in our restricted geometry and shows that it cascades along the dispersion relations during the early phase with an increasing broadening around the dispersion relations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Minimal multi-element stochastic collocation for uncertainty quantification of discontinuous functions

We propose a multi-element stochastic collocation method that can be applied in high-dimensional parameter space for functions with discontinuities lying along manifolds of general geometries. The key feature of the method is that the parameter space is decomposed into multiple elements defined by the discontinuities and thus only the minimal number of elements are utilized. On each of the resulting elements the function is smooth and can be approximated using high-order methods with fast convergence properties. The decomposition strategy is in direct contrast to the traditional multi-element approaches which define the sub-domains by repeated splitting of the axes in the parameter space. Such methods are more prone to the curse-of-dimensionality because of the fast growth of the number of elements caused by the axis based splitting. The present method is a two-step approach. Firstly a discontinuity detector is used to partition parameter space into disjoint elements in each of which the function is smooth. The detector uses an efficient combination of the high-order polynomial annihilation technique along with adaptive sparse grids, and this allows resolution of general discontinuities with a smaller number of points when the discontinuity manifold is low-dimensional. After partitioning, an adaptive technique based on the least orthogonal interpolant is used to construct a generalized Polynomial Chaos surrogate on each element. The adaptive technique reuses all information from the partitioning and is variance-suppressing. We present numerous numerical examples that illustrate the accuracy, efficiency, and generality of the method. When compared against standard locally-adaptive sparse grid methods, the present method uses many fewer number of collocation samples and is more accurate.

97 MATHEMATICS AND COMPUTING↗

RingX: Scalable Parallel Attention for Long-Context Learning on HPC

The attention mechanism has become foundational for remarkable AI breakthroughs since the introduction of the Transformer, driving the demand for increasingly longer context to power frontier models such as large-scale reasoning language models and high-resolution image/video generators. However, its quadratic computational and memory complexities present substantial challenges. Current state-of-the-art parallel attention methods, such as ring attention, are widely adopted for long-context training but utilize a point-to-point communication strategy that fails to fully exploit the capabilities of modern HPC network architectures. In this work, we propose ringX, a scalable family of parallel attention methods optimized explicitly for HPC systems. By enhancing workload partitioning, refining communication patterns, and improving load balancing, ringX achieves up to 3.4 × speedup compared to conventional ring attention on the Frontier supercomputer. Optimized for both bi-directional and causal attention mechanisms, ringX demonstrates its effectiveness through training benchmarks of a Vision Transformer (ViT) on a climate dataset and a Generative Pre-Trained Transformer (GPT) model, Llama3 8B. Our method attains an end-to-end training speedup of approximately 1.5 × in both scenarios. To our knowledge, the achieved 38% model FLOPs utilization (MFU) for training Llama3 8B with a 1M-token sequence length on 4,096 GPUs represents one of the highest training efficiencies reported for long-context learning on HPC systems. Our code implementation is available at https://github.com/jqyin/ringX-attention.

Yin, Junqi [ORNL] (ORCID:0000000338435520)↗

Adsorbate Partition Functions via Phase Space Integration: Quantifying the Effect of Translational Anharmonicity on Thermodynamic Properties

A new method for computing anharmonic thermophysical properties for adsorbates on metal surfaces is presented. Classical Monte Carlo phase space integration is performed to calculate the partition function for the motion of a hydrogen atom on Cu(111). Here, a minima-preserving neural network potential energy surface is used within the integration routine. Two different sampling schema for generating the training data are presented, and two different density functionals are used. The results are benchmarked against direct state counting results by using discrete variable representation. The phase space integration results are in excellent quantitative agreement with the benchmark results. Additionally, both the discrete variable representation and the phase space integration results confirm that the motion of H on Cu(111) is highly anharmonic. The results were applied to calculate the free energy of dissociative adsorption of H 2 and the resulting Langmuir isotherms at 400, 800, and 1200 K in a partial pressure range of 0–1 bar. It shows that the anharmonic effects lead to significantly higher predicted surface site fractions of hydrogen.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

High‐fidelity detection of crop biomass quantitative trait loci from low‐cost imaging in the field

Abstract Field‐based, rapid, and nondestructive techniques for assessing plant productivity are needed to accelerate the discovery of genotype‐to‐phenotype relationships in next‐generation biomass grass crops. The use of hemispherical imaging and light attenuation modeling was evaluated against destructive harvest measures with respect to their ability to accurately capture phenotypic and genotypic relationships in a field‐grown grass crop. Plant area index ( PAI ) estimated from below‐canopy hemispherical images, as well as a suite of thirteen traits assessed by manual destructive harvests, were measured in a Setaria recombinant inbred line mapping population segregating for aboveground productivity and architecture. A significant correlation was observed between PAI and biomass production across the population at maturity ( r 2 = .60), as well as for select diverse genotypes sampled repeatedly over the growing season ( r 2 = .79). Twenty‐seven quantitative trait loci ( QTL ) were detected for manually collected traits associated with biomass production. Of these, twenty‐one were found in four clusters of colocalized QTL . Analysis of image‐based estimates of PAI successfully identified all four QTL hot spots for biomass production. QTL for PAI had greater overlap with those detected for traits associated with biomass production than with those for plant architecture and biomass partitioning. Hemispherical imaging is an affordable and scalable method, which demonstrates how high‐throughput phenotyping can identify QTL related to biomass production of field trials in place of destructive harvests that are labor, time, and material intensive.

09 BIOMASS FUELS↗

A general asymptotic formula for distinct partitions

Many asymptotic formulas exist for unrestricted integer partitions as well as for equal partitions of integers into a finite number of parts. We use an analogy with fermion gases and the tools of statistical physics to derive asymptotic formulas for distinct partitions with a large but finite number of parts. These results are supported by the fact that we recover some other existing asymptotic results and by numerical comparisons with exact results.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

An overlapping Domain Decomposition preconditioning method for monolithic solution of shear bands

Metals subjected to high strain rate impact often exhibit a sudden and profound drop in the material’s load bearing capability, a ductile failure phenomenon known as shear banding. Shear bands, characterized as material instabilities, are driven by shear heating and described as narrow regions that have sustained intense plastic deformation and high temperature rise. This coupled thermo-mechanical localization problem can be formulated as a nonlinear system with two balance equations, momentum and energy, and two constitutive laws for elasticity and plasticity. Here in our formulation, mixed finite elements are used to discretize the equations in space and an implicit finite difference scheme is used to advance the system in time, where at every time step, a Newton type method is used to solve the system monolithically. To that end, a block Jacobian matrix is formed analytically using Gâteaux derivatives. The resulting Jacobian is sparse, nonsymmetric and its sparsity pattern and eigenvalue content vary with the different stages of shear bands formation, consequently posing a significant challenge to iterative solvers. To address this issue, an overlapping Domain Decomposition preconditioner that takes into account the physics of shear bands and the domain in which it forms, is proposed. The key idea is to resolve the shear band domain accurately while only solving the remaining domain approximately, with selective updates when necessary. The preconditioner is formed on the basis of an additive Schwartz method that is applied to a Schur complement partition of the system. The proposed preconditioner is implemented in parallel and tested on three different benchmark problems as compared against off-the-shelf solvers. We study the effect of h- and k-refinement that are obtained from Isogeometric discretizations of the system, the overlapping strategy and its parallel performance. Excellent performance is demonstrated in serial and parallel on all benchmark examples.

42 ENGINEERING↗

Automated Computational Thermochemistry for Butane Oxidation: A Prelude to Predictive Automated Combustion Kinetics

arge-scale implementation of high level computational theoretical chemical kinetics offers the prospect for dramatically improving the fidelity of combustion chemical modeling. As a first step toward this goal, we developed a procedure for automatically generating the thermochemical data for combustion of an arbitrary fuel. The procedure begins by producing a list of combustion relevant species from a specification of the fuel and combustion conditions of interest. Then, for each element in the list of species, the procedure determines an internal coordinate z-matrix description of its structure, the optimal torsional configuration via Monte Carlo sampling, key rovibrational properties for that optimal geometry (including anharmonic corrections from torsional mappings and/or vibrational perturbation theory), and high level estimates of the electronic and zero-point energies via arbitrarily defined composite methods. This dataset is then converted first to partition functions, then to thermodynamic properties, and finally to NASA polynomial representations of the data. The end product is an automatically generated database of electronic structure results and thermochemical data including representations in a format appropriate for combustion simulations. The utility and functioning of this predictive automated computational thermochemistry (PACT) software package is illustrated through application to the automated generation of thermochemical data for the combustion of n-butane. Butane is chosen for this demonstration as its species list is of reasonably manageable size for debugging level computations, while still presenting most of the key challenges that need to be surmounted in the consideration of larger fuels. Furthermore, its low temperature chemistry is representative of that oc-curring with larger alkanes.

Keceli, Murat↗

Non-pairwise additivity of the leading-order dispersion energy

The leading-order (i.e., dipole-dipole) dispersion energy is calculated for one-dimensional (1D) and two-dimensional (2D) infinite lattices, and an infinite 1D array of infinitely long lines, of doubly occupied locally harmonic wells. The dispersion energy is decomposed into pairwise and non-pairwise additive components. By varying the force constant and separation of the wells, the non-pairwise additive contribution to the dispersion energy is shown to depend on the overlap of density between neighboring wells. As well separation is increased, the non-pairwise additivity of the dispersion energy decays. The different rates of decay for 1D and 2D lattices of wells is explained in terms of a Jacobian effect that influences the number of nearest neighbors. For an array of infinitely long lines of wells spaced 5 bohrs apart, and an inter-well spacing of 3 bohrs within a line, the non-pairwise additive component of the leading-order dispersion energy is −0.11 kJ mol{sup −1} well{sup −1}, which is 7% of the total. The polarizability of the wells and the density overlap between them are small in comparison to that of the atomic densities that arise from the molecular density partitioning used in post-density-functional theory (DFT) damped dispersion corrections, or DFT-D methods. Therefore, the nonadditivity of the leading-order dispersion observed here is a conservative estimate of that in molecular clusters.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Nuclear cusps and singularities in the nonadditive kinetic potential bifunctional from analytical inversion

The non-additive kinetic potential $v^{\text{NAD}}$ is a key quantity in density-functional theory (DFT) embedding methods, such as frozen density embedding theory and partition DFT. $v^{\text{NAD}}$ is a bi-functional of electron densities $\rho_{\rm B}$ and $\rho_{\rm tot} = \rho_{\rm A} + \rho_{\rm B}$. It can be evaluated using approximate kinetic-energy functionals, but accurate approximations are challenging. The behavior of $v^{\text{NAD}}$ in the vicinity of the nuclei has long been questioned, and singularities were seen in some approximate calculations. In this article, the existence of singularities in $v^{\text{NAD}}$ is analyzed analytically for various choices of $\rho_{\rm B}$ and $\rho_{\rm tot}$, using the nuclear cusp conditions for the density and Kohn-Sham potential. It is shown that no singularities arise from smoothly partitioned ground-state Kohn-Sham densities. We confirm this result by numerical calculations on diatomic test systems HeHe, HeLi$^+$, and H$_2$, using analytical inversion to obtain a numerically exact $v^{\rm NAD}$ for the local density approximation. We examine features of $v^{\rm NAD}$ which can be used for development and testing of approximations to $v^{\rm NAD}[\rho_{\rm B},\rho_{\rm tot}]$ and kinetic-energy functionals.

74 ATOMIC AND MOLECULAR PHYSICS↗

Materials and structural aspects of advanced gas-turbine helicopter engines

Advances in materials, coatings, turbine cooling technology, structural and design concepts, and component-life prediction of helicopter gas-turbine-engine components are presented. Stationary parts including the inlet particle separator, the front frame, rotor tip seals, vanes and combustors and rotating components - compressor blades, disks, and turbine blades - are discussed. Advanced composite materials are considered for the front frame and compressor blades, prealloyed powder superalloys will increase strength and reduce costs of disks, the oxide dispersion strengthened alloys will have 100C higher use temperature in combustors and vanes than conventional superalloys, ceramics will provide the highest use temperature of 1400C for stator vanes and 1370C for turbine blades, and directionally solidified eutectics will afford up to 50C temperature advantage at turbine blade operating conditions. Coatings for surface protection at higher surface temperatures and design trends in turbine cooling technology are discussed. New analytical methods of life prediction such as strain gage partitioning for high temperature prediction, fatigue life, computerized prediction of oxidation resistance, and advanced techniques for estimating coating life are described.

Freche, J. C.↗

Parallel Gaussian elimination of a block tridiagonal matrix using multiple microcomputers

The solution of a block tridiagonal matrix using parallel processing is demonstrated. The multiprocessor system on which results were obtained and the software environment used to program that system are described. Theoretical partitioning and resource allocation for the Gaussian elimination method used to solve the matrix are discussed. The results obtained from running 1, 2 and 3 processor versions of the block tridiagonal solver are presented. The PASCAL source code for these solvers is given in the appendix, and may be transportable to other shared memory parallel processors provided that the synchronization outlines are reproduced on the target system.

Blech, Richard A.↗