Search NASA⌕ Search

SEARCH · Search NASA

Results for “partitioning algorithms”

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 361 records · Page 20

A study of fluid-structure interaction on overlapping grids with a focus on flexible beams

Fluid-structure interaction problems appear in many areas of science and engineering. Adequately understanding these problems is crucial to the study of such disciplines as aeroelastic flutter, wind turbine performance, implosions and explosions, and hemodynamics, to name a few. One technique to explore these problems is the use of numerical simulations. Although methods for the accurate and efficient numerical simulation of many problems involving only flexible structures or those involving only fluids have existed for a long time, the simulation of coupled problems, in which a flexible structure (or structures) interact with fluids remains challenging. The source of the difficulty with these problems is the construction of schemes for the coupling between the fluid and the solid that are both accurate and stable. There are two general approaches to designing a coupling procedure. In the first, the so-called monolithic approach, a single solver is implemented that handles both the fluid equations and the solid equations in a single framework. In this scheme, the solver handles the fluid and solid equations simultaneously, resulting in a strong coupling between the two. The second approach, named the partitioned approach, handles the fluid and structure domains separately. Whereas in the monolithic scheme there is a single solver for both the fluid and the structure, in the partitioned scheme there are separate solvers for the fluid and the structure. This separation heavily restricts the communication and coordination that can occur between the two solvers. Typically the only communication done in these schemes is through the application of boundary conditions (where the fluid applies a boundary condition on the fluid, and the solid one (or more) on the fluid). This makes the implementation of partitioned schemes easy, especially because they can take advantage of existing algorithms and software developed for the simulation of fluid flow or computational mechanics. However, they have a weakness – accuracy and stability are much harder to achieve. Stability is a particular problem for these schemes, as they are prone to the so-called added-mass effect.

97 MATHEMATICS AND COMPUTING↗

Pattern recognition in the satellite temperature retrieval problem

Pattern recognition procedures have been developed in order to improve the first-guess fields for satellite temperature retrievals. The first procedure is used to select one or more historical radiosonde temperature profiles as analog estimates of ambient thermal structure. The second procedure is used to organize a priori data into shape-coherent pattern libraries using structural information inherent in the data itself. On the basis of independent tests of about 800 temperature retrievals, it was found that: (1) the pattern recognition techniques reduced first-guess profile errors by nearly 50 percent in comparison with traditional partitioning schemes; and (2) with regression and physical-iterative retrieval algorithms, however, the effect of pattern recognition on temperature retrieval error was insignificant. Analysis of individual retrieval errors showed that poor retrievals may outweigh the potential benefits of both pattern recognition techniques.

Thompson, O. E.↗

Mathematical model partitioning and packing for parallel computer calculation

This paper deals with the development of multiprocessor simulations from a serial set of ordinary differential equations describing a physical system. The identification of computational parallelism within the model equations is discussed. A technique is presented for identifying this parallelism and for partitioning the equations for parallel solution on a multiprocessor. Next, an algorithm which packs the equations into a minimum number of processors is described. The results of applying the packing algorithm to a turboshaft engine model are presented.

Arpasi, Dale J.↗

Orbit determination by solving for gravity parameters with multiple arc data

The orbit of a satellite that repeats in the earth fixed coordinates is determined by combining GPS tracking data from multiple arcs. The satellite dynamics are modeled with the epoch state and a set of parameters, called the bin parameters, that account for the effect of the local gravitational field on the satellite current state. The epoch state is specific to each arc, and the bin parameters are common to all repeat arcs. The estimation algorithm is based on the Square Root Information Filter. It involves partitioning of the measurement matrix and use of the Householder transformation to combine multiple arc data and solve for the epoch states and the bin parameters. The bin parameters can then be converted into the earth's gravitational field with a modest amount of computation.

Wu, Jiun-Tsong↗

Site partitioning for distributed redundant disk arrays

Distributed redundant disk arrays can be used in a distributed computing system or database system to provide recovery in the presence of temporary and permanent failures of single sites. In this paper, we look at the problem of partitioning the sites into redundant arrays in such way that the communication costs for maintaining the parity information are minimized. We show that the partitioning problem is NP-complete and we propose two heuristic algorithms for finding approximate solutions.

Mourad, Antoine N.↗

An exponential time 2-approximation algorithm for bandwidth

The bandwidth of a graph G on n vertices is the minimum b such that the vertices of G can be labeled from 1 to n such that the labels of every pair of adjacent vertices differ by at most b. In this paper, we present a 2-approximation algorithm for the Bandwidth problem that takes worst-case {Omicron}(1.9797{sup n}) = {Omicron}(3{sup 0.6217n}) time and uses polynomial space. This improves both the previous best 2- and 3-approximation algorithms of Cygan et al. which have an {Omicron}*(3{sup n}) and {Omicron}*(2{sup n}) worst-case time bounds, respectively. Our algorithm is based on constructing bucket decompositions of the input graph. A bucket decomposition partitions the vertex set of a graph into ordered sets (called buckets) of (almost) equal sizes such that all edges are either incident on vertices in the same bucket or on vertices in two consecutive buckets. The idea is to find the smallest bucket size for which there exists a bucket decomposition. The algorithm uses a simple divide-and-conquer strategy along with dynamic programming to achieve this improved time bound.

97 MATHEMATICS AND COMPUTING↗

Parallel adaptive wavelet collocation method for PDEs

A parallel adaptive wavelet collocation method for solving a large class of Partial Differential Equations is presented. The parallelization is achieved by developing an asynchronous parallel wavelet transform, which allows one to perform parallel wavelet transform and derivative calculations with only one data synchronization at the highest level of resolution. The data are stored using tree-like structure with tree roots starting at a priori defined level of resolution. Both static and dynamic domain partitioning approaches are developed. For the dynamic domain partitioning, trees are considered to be the minimum quanta of data to be migrated between the processes. This allows fully automated and efficient handling of non-simply connected partitioning of a computational domain. Dynamic load balancing is achieved via domain repartitioning during the grid adaptation step and reassigning trees to the appropriate processes to ensure approximately the same number of grid points on each process. The parallel efficiency of the approach is discussed based on parallel adaptive wavelet-based Coherent Vortex Simulations of homogeneous turbulence with linear forcing at effective non-adaptive resolutions up to 2048{sup 3} using as many as 2048 CPU cores.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A parallel row-based algorithm for standard cell placement with integrated error control

A new row-based parallel algorithm for standard-cell placement targeted for execution on a hypercube multiprocessor is presented. Key features of this implementation include a dynamic simulated-annealing schedule, row-partitioning of the VLSI chip image, and two novel approaches to control error in parallel cell-placement algorithms: (1) Heuristic Cell-Coloring; (2) Adaptive Sequence Length Control.

Sargent, Jeff S.↗

Framework for Extensible, Asynchronous Task Scheduling (FEATS) in Fortran

Most parallel scientific programs contain compiler directives (pragmas) such as those from OpenMP, explicit calls to runtime library procedures such as those implementing the Message Passing Interface (MPI), or compiler-specific language extensions such as those provided by CUDA. By contrast, the recent Fortran standards empower developers to express parallel algorithms without directly referencing lower-level parallel programming models. Fortran’s parallel features place the language within the Partitioned Global Address Space (PGAS) class of programming models. When writing programs that exploit data-parallelism, application developers often find it straightforward to develop custom parallel algorithms. Problems involving complex, heterogeneous, staged calculations, however, pose much greater challenges. Such applications require careful coordination of tasks in a manner that respects dependencies prescribed by a directed acyclic graph. When rolling one’s own solution proves difficult, extending a customizable framework becomes attractive. The paper presents the design, implementation, and use of the Framework for Extensible Asynchronous Task Scheduling (FEATS), which we believe to be the first task-scheduling tool written in modern Fortran. We describe the benefits and compromises associated with choosing Fortran as the implementation language, and we propose ways in which future Fortran standards can best support the use case in this paper.

Richardson, Brad↗

Framework for Extensible, Asynchronous Task Scheduling (FEATS) in Fortran

Most parallel scientific programs contain compiler directives (pragmas) such as those from OpenMP, explicit calls to runtime library procedures such as those implementing the Message Passing Interface (MPI), or compiler-specific language extensions such as those provided by CUDA. By contrast, the recent Fortran standards empower developers to express parallel algorithms without directly referencing lower-level parallel programming models. Fortran’s parallel features place the language within the Partitioned Global Address Space (PGAS) class of programming models. When writing programs that exploit data-parallelism, application developers often find it straightforward to develop custom parallel algorithms. Problems involving complex, heterogeneous, staged calculations, however, pose much greater challenges. Such applications require careful coordination of tasks in a manner that respects dependencies prescribed by a directed acyclic graph. When rolling one’s own solution proves difficult, extending a customizable framework becomes attractive. The paper presents the design, implementation, and use of the Framework for Extensible Asynchronous Task Scheduling (FEATS), which we believe to be the first task-scheduling tool written in modern Fortran. We describe the benefits and compromises associated with choosing Fortran as the implementation language, and we propose ways in which future Fortran standards can best support the use case in this paper.

Modern Fortran↗

Normalized Cut Algorithm for Automated Assignment of Protein Domains

We present a novel computational method for automatic assignment of protein domains from structural data. At the core of our algorithm lies a recently proposed clustering technique that has been very successful for image-partitioning applications. This grap.,l-theory based clustering method uses the notion of a normalized cut to partition. an undirected graph into its strongly-connected components. Computer implementation of our method tested on the standard comparison set of proteins from the literature shows a high success rate (84%), better than most existing alternative In addition, several other features of our algorithm, such as reliance on few adjustable parameters, linear run-time with respect to the size of the protein and reduced complexity compared to other graph-theory based algorithms, would make it an attractive tool for structural biologists.

Samanta, M. P.↗

Intelligent System Partitioning for Agent-Based Security Constrained Optimal Power Flow

This project developed scalable, computationally efficient algorithms to solve realistic large-scale power system optimization problems as part of a larger series of competitions run by ARPA-E. These problems are important because the secure and reliable operation of the power grid, especially under increased uncertainty and variability, is growing increasingly challenging. The economic feasibility of the proposed methods developed by our team is quite low, considering it’s a purely software-based solution to operate power grids more efficiently. The technical effectiveness, as evidenced by our performance in the competition, balances heuristics and approximations to provide a tradeoff between speed and accuracy.

24 POWER TRANSMISSION AND DISTRIBUTION↗

A novel partitioning method for block-structured adaptive meshes

We propose a novel partitioning method for block-structured adaptive meshes utilizing the meshless Lagrangian particle concept. With the observation that an optimum partitioning has high analogy to the relaxation of a multi-phase fluid to steady state, physically motivated model equations are developed to characterize the background mesh topology and are solved by multi-phase smoothed-particle hydrodynamics. In contrast to well established partitioning approaches, all optimization objectives are implicitly incorporated and achieved during the particle relaxation to stationary state. Distinct partitioning sub-domains are represented by colored particles and separated by a sharp interface with a surface tension model. In order to obtain the particle relaxation, special viscous and skin friction models, coupled with a tailored time integration algorithm are proposed. Numerical experiments show that the present method has several important properties: generation of approximately equal-sized partitions without dependence on the mesh-element type, optimized interface communication between distinct partitioning sub-domains, continuous domain decomposition which is physically localized and implicitly incremental. Therefore it is particularly suitable for load-balancing of high-performance CFD simulations.

97 MATHEMATICS AND COMPUTING↗

Parallelized event chain algorithm for dense hard sphere and polymer systems

We combine parallelization and cluster Monte Carlo for hard sphere systems and present a parallelized event chain algorithm for the hard disk system in two dimensions. For parallelization we use a spatial partitioning approach into simulation cells. We find that it is crucial for correctness to ensure detailed balance on the level of Monte Carlo sweeps by drawing the starting sphere of event chains within each simulation cell with replacement. We analyze the performance gains for the parallelized event chain and find a criterion for an optimal degree of parallelization. Because of the cluster nature of event chain moves massive parallelization will not be optimal. Finally, we discuss first applications of the event chain algorithm to dense polymer systems, i.e., bundle-forming solutions of attractive semiflexible polymers.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Scalable in-memory RDFS closure on billions of triples.

We present an RDFS closure algorithm, specifically designed and implemented on the Cray XMT supercomputer, that obtains inference rates of 13 million inferences per second on the largest system configuration we used. The Cray XMT, with its large global memory (4TB for our experiments), permits the construction of a conceptually straightforward algorithm, fundamentally a series of operations on a shared hash table. Each thread is given a partition of triple data to process, a dedicated copy of the ontology to apply to the data, and a reference to the hash table into which it inserts inferred triples. The global nature of the hash table allows the algorithm to avoid a common obstacle for distributed memory machines: the creation of duplicate triples. On LUBM data sets ranging between 1.3 billion and 5.3 billion triples, we obtain nearly linear speedup except for two portions: file I/O, which can be ameliorated with the additional service nodes, and data structure initialization, which requires nearly constant time for runs involving 32 processors or more.

99 GENERAL AND MISCELLANEOUS↗

Partitioning problems in parallel, pipelined and distributed computing

The problem of optimally assigning the modules of a parallel program over the processors of a multiple computer system is addressed. A Sum-Bottleneck path algorithm is developed that permits the efficient solution of many variants of this problem under some constraints on the structure of the partitions. In particular, the following problems are solved optimally for a single-host, multiple satellite system: partitioning multiple chain structured parallel programs, multiple arbitrarily structured serial programs and single tree structured parallel programs. In addition, the problems of partitioning chain structured parallel programs across chain connected systems and across shared memory (or shared bus) systems are also solved under certain constraints. All solutions for parallel programs are equally applicable to pipelined programs. These results extend prior research in this area by explicitly taking concurrency into account and permit the efficient utilization of multiple computer architectures for a wide range of problems of practical interest.

Bokhari, S.↗

A parallel row-based algorithm with error control for standard-cell replacement on a hypercube multiprocessor

A new row-based parallel algorithm for standard-cell placement targeted for execution on a hypercube multiprocessor is presented. Key features of this implementation include a dynamic simulated-annealing schedule, row-partitioning of the VLSI chip image, and two novel new approaches to controlling error in parallel cell-placement algorithms; Heuristic Cell-Coloring and Adaptive (Parallel Move) Sequence Control. Heuristic Cell-Coloring identifies sets of noninteracting cells that can be moved repeatedly, and in parallel, with no buildup of error in the placement cost. Adaptive Sequence Control allows multiple parallel cell moves to take place between global cell-position updates. This feedback mechanism is based on an error bound derived analytically from the traditional annealing move-acceptance profile. Placement results are presented for real industry circuits and the performance is summarized of an implementation on the Intel iPSC/2 Hypercube. The runtime of this algorithm is 5 to 16 times faster than a previous program developed for the Hypercube, while producing equivalent quality placement. An integrated place and route program for the Intel iPSC/2 Hypercube is currently being developed.

Sargent, Jeff Scott↗