Search NASA⌕ Search

SEARCH · Search NASA

Results for “partitioned 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 145 records · Page 8

Iterative solution of large, sparse linear systems on a static data flow architecture - Performance studies

The applicability of static data flow architectures to the iterative solution of sparse linear systems of equations is investigated. An analytic performance model of a static data flow computation is developed. This model includes both spatial parallelism, concurrent execution in multiple PE's, and pipelining, the streaming of data from array memories through the PE's. The performance model is used to analyze a row partitioned iterative algorithm for solving sparse linear systems of algebraic equations. Based on this analysis, design parameters for the static data flow architecture as a function of matrix sparsity and dimension are proposed.

Reed, D. A.↗

Importance of rule groupings in verification of expert systems

This paper elaborates attempts to semiautomatically structure a CLIPS expert-system rule base into groups of related rules that carry the same type of information. Different distance metrics that capture relevant information from the rules for grouping are discussed. Two clustering algorithms that partition the rule base into groups of related rules are given. Two independent evaluation criteria are developed to measure the effectiveness of the grouping strategies. Results of an experiment with three sample rule bases are presented.

Mehrotra, Mala↗

An unsupervised feature extraction method for high dimensional image data compaction

A new on-line unsupervised feature extraction method for high-dimensional remotely sensed image data compaction is presented. This method can be utilized to solve the problem of data redundancy in scene representation by satellite-borne high resolution multispectral sensors. The algorithm first partitions the observation space into an exhaustive set of disjoint objects. Then, pixels that belong to an object are characterized by an object feature. Finally, the set of object features is used for data transmission and classification. The example results show that the performance with the compacted features provides a slight improvement in classification accuracy instead of any degradation. Also, the information extraction method does not need to be preceded by a data decompaction.

Ghassemian, Hassan↗

Parallel implicit unstructured grid Euler solvers

A mesh-vertex finite volume scheme for solving the Euler equations on triangular unstructured meshes is implemented on an MIMD (multiple instruction/multiple data stream) parallel computer. An explicit four-stage Runge-Kutta scheme is used to solve two-dimensional flow problems. A family of implicit schemes is also developed to solve these problems, where the linear system that arises at each time step is solved by a preconditioned GMRES algorithm. Two partitioning strategies are employed, one that partitions triangles and the other that partitions vertices. The choice of the preconditioner in a distributed memory setting is discussed. All the methods are compared both in terms of elapsed times and convergence rates. It is shown that the implicit schemes offer adequate parallelism at the expense of minimal sequential overhead. The use of a global coarse grid to further minimize this overhead is also investigated. The schemes are implemented on a distributed memory parallel computer, the iPSC/860.

Venkatakrishnan, V.↗

Parallel implicit unstructured grid Euler solvers

A mesh-vertex finite volume scheme for solving the Euler equations on triangular unstructured meshes is implemented on a multiple-instruction/multiple-data stream parallel computer. An explicit four-stage Runge-Kutta scheme is used to solve two-dimensional flow problems. A family of implicit schemes is also developed to solve these problems, where the linear system that arises at each time step is solved by a preconditioned GMRES algorithm. Two partitioning strategies are employed: one that partitions triangles and the other that partitions vertices. The choice of the preconditioner in a distributed memory setting is discussed. All of the methods are compared both in terms of elapsed times and convergence rates. It is shown that the implicit schemes offer adequate parallelism at the expense of minimal sequential overhead. The use of a global coarse grid to further minimize this overhead is also investigated. The schemes are implemented on a distributed memory parallel computer, the Intel iPSC/860.

TRT-THEORETICAL↗

Fast structural design and analysis via hybrid domain decomposition on massively parallel processors

A hybrid domain decomposition framework for static, transient and eigen finite element analyses of structural mechanics problems is presented. Its basic ingredients include physical substructuring and /or automatic mesh partitioning, mapping algorithms, 'gluing' approximations for fast design modifications and evaluations, and fast direct and preconditioned iterative solvers for local and interface subproblems. The overall methodology is illustrated with the structural design of a solar viewing payload that is scheduled to fly in March 1993. This payload has been entirely designed and validated by a group of undergraduate students at the University of Colorado using the proposed hybrid domain decomposition approach on a massively parallel processor. Performance results are reported on the CRAY Y-MP/8 and the iPSC-860/64 Touchstone systems, which represent both extreme parallel architectures. The hybrid domain decomposition methodology is shown to outperform leading solution algorithms and to exhibit an excellent parallel scalability.

Farhat, Charbel↗

Performance Comparison of a Set of Periodic and Non-Periodic Tridiagonal Solvers on SP2 and Paragon Parallel Computers

Various tridiagonal solvers have been proposed in recent years for different parallel platforms. In this paper, the performance of three tridiagonal solvers, namely, the parallel partition LU algorithm, the parallel diagonal dominant algorithm, and the reduced diagonal dominant algorithm, is studied. These algorithms are designed for distributed-memory machines and are tested on an Intel Paragon and an IBM SP2 machines. Measured results are reported in terms of execution time and speedup. Analytical study are conducted for different communication topologies and for different tridiagonal systems. The measured results match the analytical results closely. In addition to address implementation issues, performance considerations such as problem sizes and models of speedup are also discussed.

Sun, Xian-He↗

Energy Usage in an Embedded Space Vision Application on a Tiled Architecture

The need for greater autonomy in platforms such as planetary rovers is driving rapidly to codes that far overwhelm the capabilities of conventional space-qualified single core processors to run them in real-time. However, a new generation of potentially space-qualified 2D "tiled" multi-core microprocessor chips is emerging with significant performance potential. Leveraging such inherently parallel hardware for space platforms requires consideration of both time and power limitations - the latter of which is not normally done in conventional parallel computing. This paper takes one such application, Rockster, and analyzes it for energy usage when ported to a multi-core tiled chip such as may come from the Maestro program. The results demonstrate not only the criticality of memory and interconnect in the energy of real-time parallel codes, but also the effects of possible "energy-aware" changes in partitioning and algorithm design.

multi-core processors↗

Machine and Application Aware Partitioning for Adaptive Mesh Refinement Applications. In: HPDC '17 Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing

Load balancing and partitioning are critical when it comes to parallel computations. Popular partitioning strategies based on space filling curves focus on equally dividing work. The partitions produced are independent of the architecture or the application. Given the ever-increasing relative cost of data movement and increasing heterogeneity of our architectures, it is no longer sufficient to only consider an equal partitioning of work. Minimizing communication costs are equally if not more important. Our hypothesis is that an unequal partitioning that minimizes communication costs significantly can scale and perform better than conventional equal-work partitioning schemes. This tradeoff is dependent on the architecture as well as the application. We validate our hypothesis in the context of a finite-element computation utilizing adaptive mesh-refinement. Our central contribution is a new partitioning scheme that minimizes the overall runtime of subsequent computations by performing architecture and application-aware non-uniform work assignment in order to decrease time to solution, primarily by minimizing data-movement. We evaluate our algorithm by comparing it against standard space-filling curve based partitioning algorithms and observing time-to-solution as well as energy-to-solution for solving Finite Element computations on adaptively refined meshes. We demonstrate excellent scalability of our new partition algorithm up to $262,144$ cores on ORNL's Titan and demonstrate that the proposed partitioning scheme reduces overall energy as well as time-to-solution for application codes by up to 22.0%.

Fernando, Milinda↗

Fast shared-memory streaming multilevel graph partitioning

In this report we show that a fast parallel graph partitioner can benefit many applications by reducing data transfers. The online methods for partitioning graphs have to be fast and they often rely on simple one-pass streaming algorithms, while the offline methods for partitioning graphs contain more involved algorithms and the most successful methods in this category belong to the multilevel approaches. In this work, we assess the feasibility of using streaming graph partitioning algorithms within the multilevel framework. Our end goal is to come up with a fast parallel offline multilevel partitioner that can produce competitive cutsize quality. We rely on a simple but fast and flexible streaming algorithm throughout the entire multilevel framework. This streaming algorithm serves multiple purposes in the partitioning process: a clustering algorithm in the coarsening, an effective algorithm for the initial partitioning, and a fast refinement algorithm in the uncoarsening. Its simple nature also lends itself easily for parallelization. The experiments on various graphs show that our approach is on the average up to 5.1x faster than the multi-threaded MeTiS, which comes at the expense of only 2x worse cutsize.

97 MATHEMATICS AND COMPUTING↗

Computational strategy for quantifying human pesticide exposure based upon a saliva measurement

The National Research Council of the National Academies report, Toxicity Testing in the 21st Century: A Vision and Strategy, highlighted the importance of quantitative exposure data for evaluating human toxicity risk and noted that biomonitoring is a critical tool for quantitatively evaluating exposure from both environmental and occupational settings. Direct measurement of chemical exposures using personal monitoring provides the most accurate estimation of a subject’s true exposure, and non-invasive methods have also been advocated for quantifying the pharmacokinetics and bioavailability of drugs and xenobiotics. In this regard, there is a need to identify chemicals that are readily cleared in saliva at concentrations that can be quantified to support the implementation of this approach.. The current manuscript describes the use of computational modeling approaches that are closely coupled to in vivo and in vitro experiments to predict salivary uptake and clearance of xenobiotics. The primary mechanism by which xenobiotics leave the blood and enter saliva is thought to involve paracellular transport, passive transcellular diffusion, or trancellular active transport with the majority of drugs and xenobiotics cleared from plasma into saliva by passive diffusion. The transcellular or paracellular diffusion of unbound chemicals in plasma to saliva has been computational modeled using a combination of compartmental and physiologically based approaches. Of key importance for determining the plasma:saliva partitioning was the utilization of a modified Schmitt algorithm that calculates partitioning based upon the tissue composition, pH, chemical pKa and plasma protein-binding. Sensitivity analysis of key model parameters specifically identified that both protein-binding and pKa (for weak acids and bases) had the most significant impact on the determination of partitioning and that there were clear species dependent differences based upon physiological variance between rats and humans. Ongoing efforts are focused on extending this modeling strategy to an in vitro salivary acinar cell based system that will be utilized to experimentally determine and computationally predict salivary gland uptake and clearance for a broad range of xenobiotics. Hence, it is envisioned that a combination of salivary biomonitoring and computational modeling will enable the non-invasive measurement of both environmental and occupational exposure in human populations using saliva.

compuational strategy↗

Recent advances in graph partitioning

We survey recent trends in practical algorithms for balanced graph partitioning, point to applications and discuss future research directions.

97 MATHEMATICS AND COMPUTING↗

Automatic partitioning of unstructured meshes for the parallel solution of problems in computational mechanics

Most of the recently proposed computational methods for solving partial differential equations on multiprocessor architectures stem from the 'divide and conquer' paradigm and involve some form of domain decomposition. For those methods which also require grids of points or patches of elements, it is often necessary to explicitly partition the underlying mesh, especially when working with local memory parallel processors. In this paper, a family of cost-effective algorithms for the automatic partitioning of arbitrary two- and three-dimensional finite element and finite difference meshes is presented and discussed in view of a domain decomposed solution procedure and parallel processing. The influence of the algorithmic aspects of a solution method (implicit/explicit computations), and the architectural specifics of a multiprocessor (SIMD/MIMD, startup/transmission time), on the design of a mesh partitioning algorithm are discussed. The impact of the partitioning strategy on load balancing, operation count, operator conditioning, rate of convergence and processor mapping is also addressed. Finally, the proposed mesh decomposition algorithms are demonstrated with realistic examples of finite element, finite volume, and finite difference meshes associated with the parallel solution of solid and fluid mechanics problems on the iPSC/2 and iPSC/860 multiprocessors.

Farhat, Charbel↗

Hierarchical image feature extraction by an irregular pyramid of polygonal partitions

We present an algorithmic framework for hierarchical image segmentation and feature extraction. We build a successive fine-to-coarse hierarchy of irregular polygonal partitions of the original image. This multiscale hierarchy forms the basis for object-oriented image analysis. The framework incorporates the Gestalt principles of visual perception, such as proximity and closure, and exploits spectral and textural similarities of polygonal partitions, while iteratively grouping them until dissimilarity criteria are exceeded. Seed polygons are built upon a triangular mesh composed of irregular sized triangles, whose spatial arrangement is adapted to the image content. This is achieved by building the triangular mesh on the top of detected spectral discontinuities (such as edges), which form a network of constraints for the Delaunay triangulation. The image is then represented as a spatial network in the form of a graph with vertices corresponding to the polygonal partitions and edges reflecting their relations. The iterative agglomeration of partitions into object-oriented segments is formulated as Minimum Spanning Tree (MST) construction. An important characteristic of the approach is that the agglomeration of polygonal partitions is constrained by the detected edges; thus the shapes of agglomerated partitions are more likely to correspond to the outlines of real-world objects. The constructed partitions and their spatial relations are characterized using spectral, textural and structural features based on proximity graphs. The framework allows searching for object-oriented features of interest across multiple levels of details of the built hierarchy and can be generalized to the multi-criteria MST to account for multiple criteria important for an application.

99 GENERAL AND MISCELLANEOUS↗

Sampling from the Thermal Quantum Gibbs State and Evaluating Partition Functions with a Quantum Computer

We present a quantum algorithm to prepare the thermal Gibbs state of interacting quantum systems. This algorithm sets a universal upper bound D{sup {alpha}} on the thermalization time of a quantum system, where D is the system's Hilbert space dimension and {alpha}<=(1/2) is proportional to the Helmholtz free energy density. We also derive an algorithm to evaluate the partition function of a quantum system in a time proportional to the system's thermalization time and inversely proportional to the targeted accuracy squared.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

A variable multi-step method for transient heat conduction

A variable explicit time integration algorithm is developed for unsteady diffusion problems. The algorithm uses nodal partitioning and allows the nodal groups to be updated with different time steps. The stability of the algorithm is analyzed using energy methods and critical time steps are found in terms of element eigenvalues with no restrictions on element types. Several numerical examples are given to illustrate the accuracy of the method.

Smolinski, Patrick↗

Improved parallel data partitioning by nested dissection with applications to information retrieval.

The computational work in many information retrieval and analysis algorithms is based on sparse linear algebra. Sparse matrix-vector multiplication is a common kernel in many of these computations. Thus, an important related combinatorial problem in parallel computing is how to distribute the matrix and the vectors among processors so as to minimize the communication cost. We focus on minimizing the total communication volume while keeping the computation balanced across processes. In [1], the first two authors presented a new 2D partitioning method, the nested dissection partitioning algorithm. In this paper, we improve on that algorithm and show that it is a good option for data partitioning in information retrieval. We also show partitioning time can be substantially reduced by using the SCOTCH software, and quality improves in some cases, too.

99 GENERAL AND MISCELLANEOUS↗