Search NASA⌕ Search

SEARCH · Search NASA

Results for “Decomposition algorithm”

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 271 records · Page 15

QuadSync: Quadrifocal tensor synchronization via Tucker decomposition

In structure from motion, quadrifocal tensors capture more information than their pairwise counterparts (essential matrices), yet they have often been thought of as impractical and only of theoretical interest. In this work, we challenge such beliefs by providing a new framework to recover n cameras from the corresponding collection of quadrifocal tensors. We form the block quadrifocal tensor and show that it admits a Tucker decomposition whose factor matrices are the stacked camera matrices, and which thus has a multilinear rank of (4,4,4,4) independent of n. We develop the first synchronization algorithm for quadrifocal tensors, using Tucker decomposition, alternating direction method of multipliers, and iteratively reweighted least squares. We further establish relationships between the block quadrifocal, trifocal, and bifocal tensors, and introduce an algorithm that jointly synchronizes these three entities. Numerical experiments demonstrate the effectiveness of our methods on modern datasets, indicating the potential and importance of using higher-order information in synchronization.

Miao, Daniel [University of Minnesota]↗

Application of the Parareal Algorithm for Acoustic Wave Propagation

We present an application of the parareal algorithm to solve wave propagation problems in the time domain. The parareal algorithm is based on a decomposition of the integration time interval in time slices. It involves a serial prediction step based on a coarse approximation, and a correction step (computed in parallel) based on a fine approximation within each time slice. In our case, the spatial discretization is based on a spectral element approximation which allows flexible and accurate wave simulations in complex geological media. Fully explicit time advancing schemes are classically used for both coarse and fine solvers.In a first stage, we solve the 1D acoustic wave equation in an homogeneous medium in order to test stability and convergence properties of the parareal algorithm. We confirmed the stability problems outlined by Bal and Farhat et al. for hyperbolic problems. These stability issues are mitigated by a time-discontinuous Galerkin discretization of the coarse solver. It may also involve a coarser spatial discretization (hp-refinement) which helps to preserve stability and allows more significant computer savings. Besides, we explore the contribution of elastodynamic homogenization to build consistent coarse grid solvers. Extension to 2D/3D realistic geological media is an ongoing work.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Near real-time ORM measurements and SVD matrix generation for 10 Hz global orbit feedback in RHIC

To reduce the effect of trajectory perturbations ({approx}10 Hz) due to vibrations of the final focusing quadrupoles at RHIC, global orbit feedback was successfully prototyped during run-10. After upgraded to a system with 36 BPMs and 12 correctors, 10 Hz feedback was tested successfully in Run-11 and is in operational status for physics program. The test and operation of the system has been performed using transfer functions between the beam position monitors and correctors obtained from the online optics model and a correction algorithm based on singular value decomposition (SVD). One of our goals is to self-calibrate the system using SVD matrices derived from orbit response matrix (ORM) measurements acquired real-time using the new FPGA-based signal processing. Comparisons between measurement matrix and model matrix and the generation of SVD matrix for the feedback operation are presented.

43 PARTICLE ACCELERATORS↗

Limited-memory adaptive snapshot selection for proper orthogonal decomposition

Reduced order models are useful for accelerating simulations in many-query contexts, such as optimization, uncertainty quantification, and sensitivity analysis. However, offline training of reduced order models can have prohibitively expensive memory and floating-point operation costs in high-performance computing applications, where memory per core is limited. To overcome this limitation for proper orthogonal decomposition, we propose a novel adaptive selection method for snapshots in time that limits offline training costs by selecting snapshots according an error control mechanism similar to that found in adaptive time-stepping ordinary differential equation solvers. The error estimator used in this work is related to theory bounding the approximation error in time of proper orthogonal decomposition-based reduced order models, and memory usage is minimized by computing the singular value decomposition using a single-pass incremental algorithm. Results for a viscous Burgers’ test problem demonstrate convergence in the limit as the algorithm error tolerances go to zero; in this limit, the full order model is recovered to within discretization error. The resulting method can be used on supercomputers to generate proper orthogonal decomposition-based reduced order models, or as a subroutine within hyperreduction algorithms that require taking snapshots in time, or within greedy algorithms for sampling parameter space.

97 MATHEMATICS AND COMPUTING↗

Intelligent load pattern modeling and denoising using improved variational mode decomposition for various calendar periods

This paper proposes a novel approach to identify the load patterns considering calendar impact and denoising process, which advances the state of art of load identification and load forecasting to a more flexible and adaptive extent by revealing the correlation between load profile and the substantial influence factors. With the rapid development and the promising application of energy intemet and grid modernization, the accurate load modeling and forecasting especially by load pattern modeling (LPM) have become more and more critical to observe and estimate load profile so as to optimize the operation efficiency of the power grid. Traditional methods mostly focus on the static LPM itself which are unable to account the non-linear and highly dynamic load fluctuations in order to balance the demand of the power grid during various calendar period (VCP) scheduling. In this paper, a load pattern modeling method for power grid considering the impact of typical calendar effects is proposed using a combination of improved variational mode decomposition (IVMD) and deep belief network (DBN) algorithm. Firstly, an index of load curve curvature (LCC) is proposed to improve the selection of value K in the conventional VMD algorithm. The value K of the intrinsic modality is determined by adopting LCC to better quantify the average instantaneous frequency of the modal decomposition of the VMD and to reduce the elevated stochasticity of k. Secondly, an optimization formulation is developed based on adaptive segmented loop decomposition of the typical correlation coefficient to suppress the edge effect in VMD decomposition, which would enable the reconstructed curve to incorporate the variabilities from the initial data. Finally, the IVMD and DBN algorithms are jointly implemented to optimize the accuracy of VCP load pattern modeling curve regression. Numerical examples have been performed with the results proving that the proposed method significantly outperforms the state-of-art in reducing the modeling error of VCP load curve, improving the overall forecasting accuracy, and reducing load dispatching cost.

deep belief network↗

An Extended Group Additivity Method for Polycyclic Thermochemistry Estimation

Automatic kinetic mechanism generation, virtual high-throughput screening, and automatic transition state search are currently trending applications requiring exploration of a large molecule space. Large-scale search requires fast and accurate estimation of molecules’ properties of interest, such as thermochemistry. Existing approaches are not satisfactory for large polycyclic molecules: considering the number of molecules being screened, quantum chemistry (even cheap density functional theory methods) can be computationally expensive, and group additivity, though fast, is not sufficiently accurate. Here, this paper provides a fast and moderately accurate alternative by proposing a polycyclic thermochemistry estimation method that extends the group additivity method with two additional algorithms: similarity match and bicyclic decomposition. It significantly reduces H f (298 K) estimation error from over 60 kcal/mol (group additivity method) to around 5 kcal/mol, C p (298 K) error from 9 to 1 cal/mol/K,and S(298 K) error from 70 to 7 cal/mol/K. This method also works well for heteroatomic polycyclics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Block-oriented modeling of superstructure optimization problems

We present a novel software framework for modeling large-scale engineered systems as mathematical optimization problems. A key motivating feature in such systems is their hierarchical, highly structured topology. Existing mathematical optimization modeling environments do not facilitate the natural expression and manipulation of hierarchically structured systems. Rather, the modeler is forced to "flatten" the system description, hiding structure that may be exploited by solvers, and obfuscating the system that the modeling environment is attempting to represent. To correct this deficiency, we propose a Python-based "block-oriented" modeling approach for representing the discrete components within the system. Our approach is an extension of the Pyomo library for specifying mathematical optimization problems. Through the use of a modeling components library, the block-oriented approach facilitates a clean separation of system superstructure from the details of individual components. This approach also naturally lends itself to expressing design and operational decisions as disjunctive expressions over the component blocks. By expressing a mathematical optimization problem in a block-oriented manner, inherent structure (e.g., multiple scenarios) is preserved for potential exploitation by solvers. In particular, we show that block-structured mathematical optimization problems can be straightforwardly manipulated by decomposition-based multi-scenario algorithmic strategies, specifically in the context of the PySP stochastic programming library. We illustrate our block-oriented modeling approach using a case study drawn from the electricity grid operations domain: unit commitment with transmission switching and N - 1 reliability constraints. Finally, we demonstrate that the overhead associated with block-oriented modeling only minimally increases model instantiation times, and need not adversely impact solver behavior. (C) 2013 Elsevier Ltd. All rights reserved.

Friedman, Z↗

Accelerating and Parallelizing Lagrangian Simulations of Mixing-Limited Reactive Transport

Recent advances in random walk particle tracking have enabled direct simulation of mixing and reactions by allowing the particles to interact with each other using a multipoint mass transfer scheme. The mass transfer scheme allows separation of mixing and spreading processes, among other advantages, but it is computationally expensive because its speed depends on the number of interacting particle pairs. This note explores methods for relieving the computational bottleneck caused by the mass transfer step, and we use these algorithms to develop a new parallel, interacting particle model. The new model is a combination of a sparse search algorithm and a novel domain decomposition scheme, both of which offer significant speedup relative to the reference case—even when they are executed serially. We combine the strengths of these methods to create a parallel particle scheme that is highly accurate and efficient with run times that scale as 1/P for a fixed number of particles, where P is the number of computational cores (equivalently, subdomains, in this work) being used. Here, the new parallel model is a significant advance because it enables efficient simulation of large particle ensembles that are needed for environmental simulations and also because it can naturally pair with parallel geochemical solvers to create a practical Lagrangian tool for simulating mixing and reactions in complex chemical systems.

54 ENVIRONMENTAL SCIENCES↗

Three-dimensional reconstruction of neutron, gamma-ray, and x-ray sources using spherical harmonic decomposition

Neutron, gamma-ray, and x-ray imaging are important diagnostic tools at the National Ignition Facility (NIF) for measuring the two-dimensional (2D) size and shape of the neutron producing region, for probing the remaining ablator and measuring the extent of the DT plasmas during the stagnation phase of Inertial Confinement Fusion implosions. Due to the difficulty and expense of building these imagers, at most only a few two-dimensional projections images will be available to reconstruct the three-dimensional (3D) sources. In this work, we present a technique that has been developed for the 3D reconstruction of neutron, gamma-ray, and x-ray sources from a minimal number of 2D projections using spherical harmonics decomposition. We present the detailed algorithms used for this characterization and the results of reconstructed sources from experimental neutron and x-ray data collected at OMEGA and NIF.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Communication-Avoiding Parallel Sparse-Dense Matrix-Matrix Multiplication

Multiplication of a sparse matrix with a dense matrix is a building block of an increasing number of applications in many areas such as machine learning and graph algorithms. However, most previous work on parallel matrix multiplication considered only both dense or both sparse matrix operands. This paper analyzes the communication lower bounds and compares the communication costs of various classic parallel algorithms in the context of sparse-dense matrix-matrix multiplication. We also present new communication-avoiding algorithms based on a 1D decomposition, called 1.5D, which - while suboptimal in dense-dense and sparse-sparse cases - outperform the 2D and 3D variants both theoretically and in practice for sparse-dense multiplication. Our analysis separates one-time costs from per iteration costs in an iterative machine learning context. Experiments demonstrate speedups up to 100x over a baseline 3D SUMMA implementation and show parallel scaling over 10 thousand cores.

Koanantakool, Penporn↗

Beam correction for multi-pass arcs in FFA@CEBAF: status update

This work examines the multi-pass steering of six electron beams in an FFA arc ranging from approximately 10.5 GeV to 22 GeV. Shown here is an algorithm based on singular value decomposition (SVD) to successfully steer all six beams through the arc given precise knowledge of all beam positions at each of one hundred and one diagnostic locations with one hundred individual corrector magnets: that is successive application of SVD to different 100 × 101 response matrices—one for each beam energy. Further, a machine learning scheme is developed which only requires knowledge of the energy-averaged beam position at each location to provide equivalent steering. Extension of this scheme to other beam optics quantities as well as transverse and longitudinal coupling is explored.

Accelerator Physics↗

True Load Balancing for Matricized Tensor Times Khatri-Rao Product

MTTKRP is the bottleneck operation in algorithms used to compute the CP tensor decomposition. For sparse tensors, utilizing the compressed sparse fibers (CSF) storage format and the CSF-oriented MTTKRP algorithms is important for both memory and computational efficiency on distributed-memory architectures. Existing intelligent tensor partitioning models assume the computational cost of MTTKRP to be proportional to the total number of nonzeros in the tensor. However, this is not the case for the CSF-oriented MTTKRP on distributed-memory architectures. We outline two deficiencies of nonzero-based intelligent partitioning models when CSF-oriented MTTKRP operations are performed locally: failure to encode processors' computational loads and increase in total computation due to fiber fragmentation. We focus on existing fine-grain hypergraph model and propose a novel vertex weighting scheme that enables this model encode correct computational loads of processors. We also propose to augment the fine-grain model by fiber nets for reducing the increase in total computational load via minimizing fiber fragmentation. In this way, the proposed model encodes minimizing the load of the bottleneck processor. In conclusion, parallel experiments with real-world sparse tensors on up to 1024 processors prove the validity of the outlined deficiencies and demonstrate the merit of our proposed improvements in terms of parallel runtimes.

97 MATHEMATICS AND COMPUTING↗

Event-Based Parareal: A data-flow based implementation of parareal

Parareal is an iterative algorithm that, in effect, achieves temporal decomposition for a time-dependent system of differential or partial differential equations. A solution is obtained in a shorter wall-clock time, but at the expense of increased compute cycles. The algorithm combines a fine solver that solves the system to acceptable accuracy with an approximate coarse solver. The critical task for the successful implementation of parareal on any system is the development of a coarse solver that leads to convergence in a small number of iterations compared to the number of time slices in the full time interval, and is, at the same time, much faster than the fine solver. Fast coarse solvers may not lead to sufficiently rapid convergence, and slow coarse solvers may not lead to significant gains even if the number of iterations to convergence is satisfactory. We find that the difficulty of meeting these conflicting demands can be substantially eased by using a data-driven, event-based implementation of parareal instead of the conventional algorithm where solver tasks are executed sequentially. For given convergence properties, the event-based approach relaxes the speed requirements on the coarse solver by a factor of , where is the number of iterations required for a converged solution. This may, for many problems, lead to an efficient parareal implementation that would otherwise not be possible or would require substantial coarse solver development.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Exploring the Interplay of Resilience and Energy Consumption for a Task-Based Partial Differential Equations Preconditioner

We discuss algorithm-based resilience to silent data corruption (SDC) in a task- based domain-decomposition preconditioner for partial differential equations (PDEs). The algorithm exploits a reformulation of the PDE as a sampling problem, followed by a solution update through data manipulation that is resilient to SDC. The implementation is based on a server-client model where all state information is held by the servers, while clients are designed solely as computational units. Scalability tests run up to ~ 51 K cores show a parallel efficiency greater than 90%. We use a 2D elliptic PDE and a fault model based on random single bit-flip to demonstrate the resilience of the application to synthetically injected SDC. We discuss two fault scenarios: one based on the corruption of all data of a target task, and the other involving the corruption of a single data point. We show that for our application, given the test problem considered, a four-fold increase in the number of faults only yields a 2% change in the overhead to overcome their presence, from 7% to 9%. We then discuss potential savings in energy consumption via dynamics voltage/frequency scaling, and its interplay with fault-rates, and application overhead.

97 MATHEMATICS AND COMPUTING↗

MatRIS: Addressing the Challenges for Portability and Heterogeneity Using Tasking for Matrix Decomposition (Cholesky)

The ubiquitous in-node heterogeneity of HPC and cloud computing platforms makes software portability and performance optimization extremely challenging. Described here, the MatRIS multilevel math library abstraction framework employs tasking to alleviate these difficulties. MatRIS includes the IRIS task-based runtime on the bottom level and exposes different layers of abstraction to render algorithms architecturally agnostic. MatRIS ensures the decomposition and creation of tasks that represent the necessary encapsulation of the optimized kernels from both vendor and open-source math libraries. Once built, MatRIS can select different combinations of accelerators at runtime, making it portable even on diverse heterogeneous architectures. By leveraging the IRIS runtime’s features for managing heterogeneity, MatRIS deploys algorithms that remove the need to specify orchestration and data transfer. This study describes how the serial task abstraction of a tiled Cholesky factorization is made portable and scalable in the case of multi-device and multi-vendor heterogeneity on a node with NVIDIA and AMD GPUs by using MatRIS. First, we demonstrate that Cholesky in MatRIS provides multi-GPU scalability that offers competitive performance versus cuSolverMG. Then, we present the challenges and opportunities for heterogeneous execution.

Monil, M. A. H.↗

Exploring the interplay of resilience and energy consumption for a task-based partial differential equations preconditioner

Here, we discuss algorithm-based resilience to silent data corruptions (SDCs) in a task-based domain-decomposition preconditioner for partial differential equations (PDEs). The algorithm exploits a reformulation of the PDE as a sampling problem, followed by a solution update through data manipulation that is resilient to SDCs. The implementation is based on a server-client model where all state information is held by the servers, while clients are designed solely as computational units. Scalability tests run up to ~51K cores show a parallel efficiency greater than 90%. We use a 2D elliptic PDE and a fault model based on random single and double bit-flip to demonstrate the resilience of the application to synthetically injected SDC. We discuss two fault scenarios: one based on the corruption of all data of a target task, and the other involving the corruption of a single data point. We show that for our application, given the test problem considered, a four-fold increase in the number of faults only yields a 2% change in the overhead to overcome their presence, from 7% to 9%. We then discuss potential savings in energy consumption via dynamic voltage/frequency scaling, and its interplay with fault-rates, and application overhead.

97 MATHEMATICS AND COMPUTING↗

Lifting MGARD: Construction of (pre)wavelets on the interval using polynomial predictors of arbitrary order

MGARD (MultiGrid Adaptive Reduction of Data) is an algorithm for compressing and refactoring scientific data, based on the theory of multigrid methods. The core algorithm is built around stable multilevel decompositions of conforming piecewise linear $C^0$ finite element spaces, enabling accurate error control in various norms and derived quantities of interest. In this work, we extend this construction to arbitrary order Lagrange finite elements $\mathbb{Q}_p$, $p \geq 0$, and propose a reformulation of the algorithm as a lifting scheme with polynomial predictors of arbitrary order. Additionally, a new formulation using a compactly supported wavelet basis is discussed, and an explicit construction of the proposed wavelet transform for uniform dyadic grids is described.

Reshniak, Viktor [Oak Ridge National Laboratory (O↗