We have developed a coarse-grain parallelization of the Buchberger algorithm for computing Grobner bases in algebras of linear differential operators. The implementation of this algorithm provides good speedups on the...
详细信息
ISBN:
(纸本)0769521983
We have developed a coarse-grain parallelization of the Buchberger algorithm for computing Grobner bases in algebras of linear differential operators. The implementation of this algorithm provides good speedups on the majority of examples coming from these noncommutative algebras, which are superior to the speedups achieved for (commutative) polynomial rings.
A management configuration refers to an arrangement of management agents across a distributed system whose tasks are to monitor and control system components to ensure efficient use of resources and timely services to...
详细信息
ISBN:
(纸本)1932415262
A management configuration refers to an arrangement of management agents across a distributed system whose tasks are to monitor and control system components to ensure efficient use of resources and timely services to users. These management activities can degrade system performance due to the resources consumed by management agents, e.g. CPU, memory, network bandwidth, storage space, etc. A good configuration of management agents is needed to minimize the impact on system performance. Previous work on determining efficient management configurations proposed a branch-and-bound based algorithm for finding a good configuration. The performance of that algorithm can be unacceptable even for moderately sized systems. A parallel version of that algorithm has been developed. This paper reports on an implementation and experiments evaluating the effectiveness of that algorithm.
Concept Lattice, the core data structure in Formal Concept Analysis, has been used widely in machine learning, data mining and knowledge discovery, information retrieval, etc. The main difficulty with concept lattice-...
详细信息
ISBN:
(纸本)0769521525
Concept Lattice, the core data structure in Formal Concept Analysis, has been used widely in machine learning, data mining and knowledge discovery, information retrieval, etc. The main difficulty with concept lattice-based system comes from the lattice construction itself In this paper, a parallel algorithm based on the closure search space partition for computing concepts is proposed. This algorithm divides the closure search space into several subspaces in accordance with criterions prescribed ahead and introduces an efficient scheme to recognize the valid ones, in which the searching for closures is bounded An intermediate structure is employed to judge the validity of a subspace and compute closures more efficiently. Since the searching in subspaces are independent tasks, a parallel algorithm based on search space partition can be directly reached.
In this paper special attention is devoted to one particular problem, the diffusion equation subject to the specification of mass in a portion of the domain, which has been studied quite extensively, both analytically...
详细信息
In this paper special attention is devoted to one particular problem, the diffusion equation subject to the specification of mass in a portion of the domain, which has been studied quite extensively, both analytically and numerically, in recent years. Parabolic partial differential equations with a non-local constraint in place of one of the standard boundary specifications feature in the mathematical modeling of many phenomena. In this paper the application of the method of lines (MOL) to such problems is considered. The MOL semi-discretization approach will be used to transform the model partial differential equation into a system of first-order linear ordinary differential equations (ODES). The MOL is a method of solving PDEs by discretizing the equation with respect to all but one variable (usually time). The spatial partial derivative is approximated by a finite-difference method. The solution of the resulting system of first-order ODEs satisfies a recurrence relation which involves a matrix exponential function. Numerical techniques are developed by approximating the exponential matrix function in this recurrence relation. The matrix exponential function is approximated by a rational approximation consisting of four parameters. New parallel algorithms are developed using the resulting approximation. Numerical experiments on two challenging examples are presented to illustrate the performance of the algorithms. (C) 2003 Elsevier Inc. All rights reserved.
A key ingredient of finite-element analysis programs is the linear-algebra solver, typically either a linear-system solver or an eigensolver. The first part of this paper tries to justify why it is important to have r...
详细信息
A key ingredient of finite-element analysis programs is the linear-algebra solver, typically either a linear-system solver or an eigensolver. The first part of this paper tries to justify why it is important to have recourse to publicly available software for addressing this part of the computation. A number of libraries are mentioned as successful examples that exhibit a series of desirable qualities. Although some of these libraries force the programmer to somewhat change the programming style and may be difficult to learn, the benefits usually pay off the extra effort. The second part of the paper describes one of these libraries in some detail, namely SLEPc, the Scalable Library for Eigenvalue Problem Computations, which is used to illustrate the benefits of modern software paradigms for scientific and engineering computing.
Using orthogonal localized occupied orbitals we have developed and implemented a parallelized local second-order Moller-Plesset (MP2) method based on the idea developed by Head-Gordon and co-workers. A subset of nonor...
详细信息
Using orthogonal localized occupied orbitals we have developed and implemented a parallelized local second-order Moller-Plesset (MP2) method based on the idea developed by Head-Gordon and co-workers. A subset of nonorthogonal correlation functions (the orbital domain) was assigned to each of the localized occupied orbitals using a distance criterion and excitations from localized occupied orbitals that were arranged into subsets. The correlation energy was estimated using a partial diagonalization and an iterative efficient method for solving large-scale linear equations. Some illustrative calculations are provided for molecules with up to 1484 Cartesian basis sets. The orbital domain sizes were found to be independent of the molecular size, and the present local MP2 method covered about 98%-99% of the correlation energy of the conventional canonical MP2 method. (C) 2004 American Institute of Physics.
We present the first molecular simulations of the vapor-liquid surface tension of quantum,liquids. The path integral formalism of Feynman was used to account for the quantum mechanical behavior of both the liquid and ...
详细信息
We present the first molecular simulations of the vapor-liquid surface tension of quantum,liquids. The path integral formalism of Feynman was used to account for the quantum mechanical behavior of both the liquid and the vapor. A replica-data parallel algorithm was implemented to achieve good parallel performance of the simulation code on at least 32 processors. We have computed the surface tension and the vapor-liquid phase diagram of pure hydrogen over the temperature range 18-30 K and pure deuterium from 19 to 34 K. The simulation results for surface tension and vapor-liquid orthobaric densities are in very good agreement with experimental data. We have computed the interfacial properties of hydrogen-deuterium mixtures over the entire concentration range at 20.4 and 24 K. The calculated equilibrium compositions of the mixtures are in excellent agreement with experimental data. The computed mixture surface tension shows negative deviations from ideal solution behavior,,in agreement with experimental data and predictions from Prigogine's theory. The magnitude of the deviations at 20.4 K are substantially larger from simulations and from theory than from experiments. We conclude that the experimentally measured mixture surface tension values are systematically too high. Analysis of the concentration profiles in the interfacial region shows that the nonideal behavior can be described entirely by segregation of H-2 to the interface, indicating that H-2 acts as a surfactant in H-2-D-2 mixtures. (C) 2004 American Institute of Physics.
Many approaches have been described for the parallel loop scheduling problem for shared-memory systems, but little work has been done on the data-dependent loop scheduling problem (nested loops with loop carried depen...
详细信息
Many approaches have been described for the parallel loop scheduling problem for shared-memory systems, but little work has been done on the data-dependent loop scheduling problem (nested loops with loop carried dependencies). In this paper, we propose a general model for the data-dependent loop scheduling problem on distributed as well as shared memory systems. In order to achieve load balancing and low runtime scheduling and communication overhead, our model is based on a loop task graph and the notion of critical path. In addition, we develop a heuristic algorithm based on our model and on genetic algorithms to test the reliability of the model. We test our approach on different scenarios and benchmarks. The results are very encouraging and suggest a future parallel compiler implementation based on our model. (C) 2004 Elsevier Inc. All rights reserved.
Clustering is a fundamental and important technique in image processing, pattern recognition, data compression, etc. However, most recent clustering algorithms cannot deal with large, complex databases and do not alwa...
详细信息
Clustering is a fundamental and important technique in image processing, pattern recognition, data compression, etc. However, most recent clustering algorithms cannot deal with large, complex databases and do not always achieve high clustering results. This paper proposes a parallel clustering algorithm for categorical and mixed data which can overcome the above problems. Our contributions are: (1) improving the k-sets algorithm [3] to achieve highly accurate clustering results;and (2) applying parallel techniques to the improved approach to achieve a parallel algorithm. Experiments on a CRAY T3E show that the proposed algorithm can achieve higher accuracy than previous attempts and can reduce processing time;thus, it is practical for use with very large and complex databases.
Watershed segmentation/transform is a classical method for image segmentation in gray scale mathematical morphology. Nevertheless watershed algorithm has strong recursive nature, so straightforward parallel one has a ...
详细信息
ISBN:
(纸本)3540233881
Watershed segmentation/transform is a classical method for image segmentation in gray scale mathematical morphology. Nevertheless watershed algorithm has strong recursive nature, so straightforward parallel one has a very low efficiency. Firstly, the advantages and disadvantages of some existing parallel algorithms are analyzed. Then, a Further Optimized parallel Watershed Algorithm (FOPWA) is presented based on boundary components graph. As the experiments show, FOPWA optimizes both running time and relative speedup, and has more flexibility.
暂无评论