In this paper, using the metric of iso--efficiency function [if. we analyze the scalability of PSRS (parallel. Sorting by Regular Sample) algorithm I2] on two popular architectures (Mesh and Hypercube) The Isoefficien...
详细信息
In this paper, using the metric of iso--efficiency function [if. we analyze the scalability of PSRS (parallel. Sorting by Regular Sample) algorithm I2] on two popular architectures (Mesh and Hypercube) The Isoefficiency function of PSRS on 2-dimensional mesh With p processors reaches the lower bound for that of sorting algorithms on this architecture. In nils sense, we say the scalabilify of PSRS is optimal on 2-dimensional mesh. The lso-efficiency function of PSRS on hypercube is equal to that of PSRS on 2-dimensional mesh. After changing the data exchanging scheme of PSRS, -cafe get a ne'v iso-efficiency function . which is better than that of PSRS on 2-dimensional mesh So we say that hypercube is more suitable for PSRS than 2--dimensional mesh.
This article describes the pack & unpack performance of the vector using YH-1 supercomputer as its platform, with its emphasis on the pack & unpack technique's unique application to vector computation, in ...
详细信息
This article describes the pack & unpack performance of the vector using YH-1 supercomputer as its platform, with its emphasis on the pack & unpack technique's unique application to vector computation, in the light of vector computer's characteristics and some practical cases. The practical numerical experiment proves that the pack & unpack techniques is a key method to the depth parallel development and vectorization research of large scale scientific computation project, and which is a highly effective method to vectorizing application programs.
This paper presents parallel algorithms for the dynamic programming problem on systolic arrays. The dynamic programming problem when implemented sequentially takes 0(n3) time steps. In this paper, an algorithm that ru...
详细信息
We consider a new supercube architecture, a new interconnection network derived from the hypercube. The supcrcube retains the connectivity and diameter properties of the corresponding hypercube. The embedding of one i...
详细信息
We consider a new supercube architecture, a new interconnection network derived from the hypercube. The supcrcube retains the connectivity and diameter properties of the corresponding hypercube. The embedding of one interconnection network into another is a very important issue in the design and analysis of parallel algorithms. In this paper, the problem of embedding and reconfiguring complete binary tree structures is considered in a supercube with faulty nodes. We also propose a new method for embedding and reconfiguring complete binary trees in a faulty supercube. Furthermore, the results enable us to obtain a good method for embedding complete binary tree structures into a faulty supercube with n-expansion. The result enables us to obtain O(n2 - m2) faults which can be tolerated, where (n - 1) is the dimension of a supercube and (m - 1) is the height of a complete binary tree.
We propose splitting, parallel algorithms for solving strongly equilibrium problems over the intersection of a finite number of closed convex sets given as the fixed-point sets of nonexpansive mappings in real Hilbert...
详细信息
We propose splitting, parallel algorithms for solving strongly equilibrium problems over the intersection of a finite number of closed convex sets given as the fixed-point sets of nonexpansive mappings in real Hilbert spaces. The algorithm is a combination between the gradient method and the Mann-Krasnosel'skii iterative scheme, where the projection can be computed onto each set separately rather than onto their intersection. Strong convergence is proved. Some special cases involving bilevel equilibrium problems with inverse strongly monotone variational inequality, monotone equilibrium constraints and maximal monotone inclusions are discussed. An illustrative example involving a system of integral equations is presented.
This paper presents a strategy to tackle the multiple sequence alignment (MSA) problem, which is one of the most important tasks in the biological sequence analysis. Its role is to align the sequences in their entiret...
详细信息
This paper presents a strategy to tackle the multiple sequence alignment (MSA) problem, which is one of the most important tasks in the biological sequence analysis. Its role is to align the sequences in their entirety to derive relationships and common characteristics between a set of protein or nucleotide sequences. The MSA problem was proved to be an NP-Hard problem. The proposed strategy incorporates a new idea based on the well-known divide-and-conquer paradigm. This paper presents a novel method of clustering sequences as a preliminary step to improve the final alignment;this decomposition can be used as an optimization procedure with any MSA aligner to explore promising alignments of the search space. In their solution, the authors proposed to align the clusters in a parallel and distributed way in order to benefit from parallel architectures. The strategy was tested using classical benchmarks like BAliBASE, Sabre, Prefab4, and Oxm, and the experimental results show that it gives good results by comparing to the other aligners.
Data currency is a temporal reference of the data, it reflects the degree to which the data is current with the world it models. Currency has a significant impact on the quality and value of the data. Once the time st...
详细信息
Data currency is a temporal reference of the data, it reflects the degree to which the data is current with the world it models. Currency has a significant impact on the quality and value of the data. Once the time stamp of the data is lost or tampered with, it is difficult to perform an absolute and precise repair. Based on data currency research, the basic currency rules were extended to support the parallelization of rule extraction and incremental updating and theoretically reduce the algorithm time complexity from O(n - 1) to O(log (n)). In practical experiments, the repair efficiency of multithreading is improved by up to 75.2% compared with single-threaded operation. According to the problems and requirements encountered in data cleaning, rulebased methods for relative currency repair and anomaly data detection were proposed, a relative currency repair algorithm was implemented, models for evaluating repair results were established, and the method of applying a repair algorithm to detect abnormal currency data was also discussed. The experimental results and analysis show that the extended currency rules that provide more valuable features are feasible and available, and the relative currency repair algorithm can effectively perform data currency repair and anomaly detection.
A new dynamic data structure has been proposed recently in 2011. There are several algorithms for matrix multiplication. But none of them has used r-train data structure for storing and multiplying the matrices. In th...
详细信息
A new dynamic data structure has been proposed recently in 2011. There are several algorithms for matrix multiplication. But none of them has used r-train data structure for storing and multiplying the matrices. In this paper algorithm for matrix multiplication using r-train for parallel machine has been proposed.
This paper presents a contour line generation parallel method based on quadratic isoparametric element. In this method, the whole area is divided into many quadratic isoparametric elements by means of finite element m...
详细信息
This paper presents a contour line generation parallel method based on quadratic isoparametric element. In this method, the whole area is divided into many quadratic isoparametric elements by means of finite element method. The contour line in each isoparametric element is generated in parallel, and the whole contour is assembled. The result shows that this method improves the computation efficiency under the condition that the precision can be guaranteed.
We are concerned with the mapping on high performance hybrid architectures of a parallel software implementing a two level overlapping domain decomposition, that is, along space and time directions, of the four dimens...
详细信息
We are concerned with the mapping on high performance hybrid architectures of a parallel software implementing a two level overlapping domain decomposition, that is, along space and time directions, of the four dimensional variational data assimilation model. The reference architecture belongs to the SCoPE (Sistema Cooperativo Per Elaborazioni scientifiche multidisciplinari) data center, located at University of Naples Federico II. We consider the initial boundary problem of the shallow water equation and analyse both strong and weak scaling. Keeping the efficiency always greater than 60%$$ 60\% $$ and about 90%$$ 90\% $$ in most cases, we experimentally find that the isoefficiency function grows a little more than linearly with respect to the number of processes. Results, obtained by using the parallel computing toolbox of MATLABR2013a, are in agreement with the algorithm's performance prevision based on the scale up factor, confirming the appropriate mapping of the algorithm on the hybrid architecture.
暂无评论