parallel computation is an effective approach to real-time simulation and transient stability online assessment of large-scale power systems. In this paper, the s-stage 2s-order symplectic Runge-Kutta-Nystrom method i...
详细信息
ISBN:
(纸本)9783037853191
parallel computation is an effective approach to real-time simulation and transient stability online assessment of large-scale power systems. In this paper, the s-stage 2s-order symplectic Runge-Kutta-Nystrom method is adopted for transient stability simulation of power system using classic model. Using Butcher transformation, a new parallel algorithm has been derived. The proposed algorithm has the convergence characteristic of a Newton type method and is of fully parallel-in-time. Through numerical simulation where the IEEE 145-bus power system is used, the proposed algorithm has been tested and compared with the conventional parallel-in-time Newton approach using implicit trapezoidal rule.
MapReduce has recently become very successful parallel processing technique. Latest DNA sequencing technologies are now able to generate huge DNA sequences easily and cheaper. Consequently making it a challenge for si...
详细信息
ISBN:
(纸本)9781509030507
MapReduce has recently become very successful parallel processing technique. Latest DNA sequencing technologies are now able to generate huge DNA sequences easily and cheaper. Consequently making it a challenge for single-core processor systems to mine patterns, hence leading to unsatisfactory performance. In this paper, we explore this challenge by making use of MapReduce on Hadoop platform using a successful data structure called the generalized suffix tree. Our experimental results show that the proposed approach can index long sequences with improved performance than previous related approaches.
GRAPES (Global/Regional Assimilation and PrEdiction System) is a new developed numerical weather prediction system and will be implemented operationally in the next few years at China Meteorological Administration (CM...
详细信息
ISBN:
(纸本)9780769541105
GRAPES (Global/Regional Assimilation and PrEdiction System) is a new developed numerical weather prediction system and will be implemented operationally in the next few years at China Meteorological Administration (CMA). For a global semi-implicit semi-Lagrangian numerical prediction model formulated in spherical coordinates, due to the convergence of meridians, the longitudinal grid size decreases toward zero as the poles are approached. Therefore, parallelism near the poles is a tough issue. With efficiency, portability, maintainability and extensibility requirements, a cap-longitude-latitude decomposition parallel algorithm is proposed and realized adherence to the architectures of high-performance computers at CMA. The results indicate that the computing performance of the proposed algorithm is good on IBM-cluster 1600 at CMA. And it can resolve effectively the occurrence of calculated zonal wind exceeding the maximum of the halo regions when locating semi-Lagrangian departure points. The algorithm is efficient and stable and can meet the operational implementation requirement.
This article considers the algorithmic aspects of parallelization of the numerical implementation of the mathematical model of anisotropic heat and mass transfer, obtained on the basis of the use of fractional order d...
详细信息
ISBN:
(纸本)9798350310856
This article considers the algorithmic aspects of parallelization of the numerical implementation of the mathematical model of anisotropic heat and mass transfer, obtained on the basis of the use of fractional order derivatives. Software for implementing a parallel algorithm for modeling heat and mass transfer in media with a fractal structure is presented. The acceleration indicators of the parallel algorithm with different discretization parameters were studied on the basis of numerical experiments. A parallel implementation was also compared with a serial solution. The proposed parallel algorithms and their software implementation can scale well on multi-core computers.
This paper initiates the studies of parallel algorithm for core maintenance in dynamic graphs. The core number is a fundamental index reflecting the cohesiveness of a graph, which is widely used in large-scale graph a...
详细信息
ISBN:
(纸本)9781538617915
This paper initiates the studies of parallel algorithm for core maintenance in dynamic graphs. The core number is a fundamental index reflecting the cohesiveness of a graph, which is widely used in large-scale graph analytics. We investigate the parallelism in the core update process when multiple edges and vertices are inserted. Specifically, we discover a structure called superior edge set, the insertion of edges in which can be processed in parallel. Based on the structure of superior edge set, an efficient parallel algorithm is then devised. To the best of our knowledge, the proposed algorithm is the first parallel one for the fundamental core maintenance problem. Finally, extensive experiments are conducted on different types of real-world and synthetic datasets, and the results illustrate the efficiency, stability and scalability of the proposed algorithm. The algorithm shows a significant speedup in the processing time compared with previous results that sequentially handle edge and vertex insertions.
The paper is devoted to the construction of an algorithm for parallelization of temperature field modeling processes based on bicubic spline and the problems that arise in it. It takes a lot of time and memory on a co...
详细信息
ISBN:
(纸本)9781665440677
The paper is devoted to the construction of an algorithm for parallelization of temperature field modeling processes based on bicubic spline and the problems that arise in it. It takes a lot of time and memory on a computer to model existing large-scale signals of temperature fields in three-dimensional graphical form. Solving these problems on the basis of the proposed parallel algorithm, the development of parallel algorithms and software tools to qualitatively determine the coordinates of high-temperature hotspots of temperature fields and allow them to be monitored continuously, and to calculate the digital processing of existing large-scale signals of temperature fields. Based on this, the article presents a parallel algorithm for digital processing of signals of temperature fields on the basis of bicubic spline and ways to parallel its computational processes using Open MP technology.
Since MODIS data have the feature of huge capacity and multi-spectrum, it needs too much time and frequent I/O operations to correct them by RS software. This paper proposes a parallel algorithm for MODIS Data based o...
详细信息
ISBN:
(纸本)0780390504
Since MODIS data have the feature of huge capacity and multi-spectrum, it needs too much time and frequent I/O operations to correct them by RS software. This paper proposes a parallel algorithm for MODIS Data based on triangulation network. The input images are divided into several pieces and each CPU processes a piece independently. By implementing the algorithm on a cluster system, the results show that, this parallel algorithm improves the efficiency of geometrical correction greatly.
In this paper, the problem of stretching a layout in the knock-knee mode is represented while considering its wirability. The previous O(n x m) time serial algorithm for this problem which is based on two layers needs...
详细信息
ISBN:
(纸本)0818673524
In this paper, the problem of stretching a layout in the knock-knee mode is represented while considering its wirability. The previous O(n x m) time serial algorithm for this problem which is based on two layers needs (2m - 1) × n layout area in the worst case, where m = number of rows and n = number of columns. However, O(n log n) time serial algorithm that is able to wire all the two-terminal nets within (m × 3n/2) layout area is given in the paper. This serial algorithm is applied to generate a parallel CREW PRAM algorithm that can wire all the nets in O(log n) time by using O(n) number of processors.
Treatment planning based on numerical prediction before or during cryosurgery is an indispensable way to achieve exactly killing of tumor. However, conventional sequential computation is difficult to meet the challeng...
详细信息
ISBN:
(纸本)9783037854693
Treatment planning based on numerical prediction before or during cryosurgery is an indispensable way to achieve exactly killing of tumor. However, conventional sequential computation is difficult to meet the challenge of real-time assistance with complex treatment plans. In this study, two parallel numerical algorithms, i.e. explicit finite difference scheme and alternating direction implicit scheme, based on an effective heat capacity method are established to solve three-dimensional phase change problems in biological tissues subjected to freezing of multiple cryoprobes. The results as well as speedup of parallel computing were compared. It was shown that the parallel algorithms developed in this study can be used to perform rapid prediction of temperature distribution for cryosurgery, and that parallel computing is hopeful to assist cryosurgeons with prospective parallel treatment planning in the near future.
This study provides a parallel algorithm for meshfree analysis based on the radial point interpolation method. This algorithm is designed to suit graphics hardware that support compute unified device architecture, NVI...
详细信息
ISBN:
(纸本)9781424470594
This study provides a parallel algorithm for meshfree analysis based on the radial point interpolation method. This algorithm is designed to suit graphics hardware that support compute unified device architecture, NVIDIA's software development environment on graphics processing units (GPUs). The radial point interpolation method is a meshfree method that enables strict imposition of Dirichlet boundary conditions, and it mainly consists of two steps: the process for assembly of the discrete linear system and the process for solving the linear system. This paper shows that both the processes can be performed effectively on the GPU by applying our parallel algorithm and at the same time, the costs for data transfer to and from the GPU can be reduced, because the processes related to the discrete linear system are performed independently on the GPU. Our numerical tests confirm that the algorithm performed on the GPU accelerates the computations by a factor of maximum 15.
暂无评论