摘要:

In this paper block-methods For second order initial value problems are developed. The matrix coefficients of the schemes are chosen in such a way as to ensure stability properties. There is anticipated speed up of computations as a result of the admissible parallelism across the method (or time) and the cheap error estimators.

ISSN:1063-7192    

DOI:10.1080/10637199408915419

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

In this paper we present a multisplitting generalized iterative method, for solving large nonsingular systems of equationsAx = b.This multisplitting iterative method contains the multisplitting GAOR method and its special cases. An analysis of its convergence is given, considering the matrixAas anH-matrix. Under the same assumption, we obtain sufficient convergence conditions for a multisplitting method, which also generalizes the method of arithmetic mean.

ISSN:1063-7192    

DOI:10.1080/10637199408915420

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

A new orthogonal decomposition for square dense matrices is proposed. For matrices of bigger size the computational time for this decomposition and the corresponding solution method is less than the computational time for the usual Givens QR decomposition The backward stability of the proposed decomposition is studied, and it is proved that the backward error depends linearly on the size of the matrix. Finally, some numerical evidence is given.

ISSN:1063-7192    

DOI:10.1080/10637199408915421

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

Efficient interprocessor communication is a key to the future success of large-scale multicomputers.Multicastcommunication, highly demanded in many scientific parallel algorithms, usually needs a large amount of communication traffic In this paper, our efforts are to alleviate the communication traffic of multicast in hypercubes. An optimal multicast algorithm in hypercubes should send a message from the source node to each of its destination nodes with a minimum number of hops while generating the least amount of total traffic Unfortunately, the problem to find optimal multicast communication in hypercubes isNP-complete. Therefore, we propose a heuristic multicast algorithm with time complexity O(nN) inn-dimensional hypercubes, whereN= 2n, This algorithm is based on the virtual-circuited mode with centralized control. Our algorithm guarantees that each of the destination nodes can receive the source message through a shortest path from the source node. Compared with other heuristic algorithm, by simulation, our algorithm is shown to reduce communication traffic in multicast communication.

ISSN:1063-7192    

DOI:10.1080/10637199408915422

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

In the past few years, there have been significant developments in the area of distributed and parallel processing. More powerful and new hardware architectures are being produced at a rapid rate, such as distributed-memory MIMD computers, which have provided enormous computing power to the software engineers. These multiprocessors may provide a significant speed-up over the serial execution of an algorithm. However, this requires careful partitioning and allocation of data and control to the processor set. Matrix multiplication is a fundamental parallel algorithm which can be effectively executed on a distributed-memory multiprocessor and can show significant improvement in the speed-up over the serial execution. Ideally, we should be able to achieve a linear speed up with increase in the number of processors, but in practice the speed up is much less, and in fact increasing the number of processors beyond a certain number may result in degradation of the completion time. This degradation is caused by increased communications between modules. Therefore, the optimum speed-up is a function of the number of processors and the communication cost. To find the optimum performance, a user need to experiment with all the available processors on a multiprocessor.

ISSN:1063-7192    

DOI:10.1080/10637199408915423

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

Adirectparallel algorithm to tridiagonalize a real symmetric matric matrix using Givens transformation is presented. The algorithm is optimal and has a time complexity ofO(NlogN) for matrices of order N.

ISSN:1063-7192    

DOI:10.1080/10637199408915424

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor

摘要:

The barotropic model is used to explore the advantages of parallel processing in deterministic forecasting. We apply this model to the track forecasting of hurricane Elena (1985).

ISSN:1063-7192    

DOI:10.1080/10637199408915425

出版商:Taylor & Francis Group    

年代:1994

数据来源: Taylor