Search results

Presents a review on implementing finite element methods on supercomputers, workstations and PCs and gives main trends in hardware and software developments. An appendix included at the end of the paper presents a bibliography on the subjects retrospectively to 1985 and approximately 1,100 references are listed.

To address the problems of high performance computing by using the networks of workstations (NOW) and to discuss the complex performance evaluation of centralised and distributed parallel algorithms.

Defines the role of performance and performance evaluation methods using a theoretical approach. Presents concrete parallel algorithms and tabulates the results of their performance.

Sees that a network of workstations based on powerful personal computers belongs in the future and as very cheap, flexible and perspective asynchronous parallel systems. Argues that this trend will produce dynamic growth in the parallel architectures based on the networks of workstations.

We would like to continue these experiments in order to derive more precise and general formulae for typical used parallel algorithms from linear algebra and other application oriented parallel algorithms.

Describes how the use of NOW can provide a cheaper alternative to traditionally used massively parallel multiprocessors or supercomputers and shows the advantages of unifying the two disciplines that are involved.

Produces a new approach and exploits the parallel processing capability of NOW. Gives the concrete practical examples of the method that has been developed using experimental measuring.

Presents a new powerful multiple‐criteria decision‐making (MCDM) framework with an approach that combines recent results from several related areas. The De Novo programming and external reconstruction approach (ERA) provides the overall structure for the algorithms and the sequence of r‐constrained linear programs generated by the ERA‐framework is solved. This is by a parallel implementation of a powerful interior point algorithm called the conjugate gradient method (CGM), selected because it is particularly suitable for parallel processing. Gives details of the proposed parallel implementation of the CGM, together with the algorithm. Presents an analysis of the parallel performance and discusses theoretical speed‐up.

Presents a framework for deciding on a good execution strategy for a given program based on the available data and task parallelism in the program on PC laboratory clusters. Proposes a virtual cluster scheduling scheme to take account of the relationships between tasks for task parallelism, and also processor speed, processor load and network environment to balance load for data parallelism in a PC cluster environment. The approach is very effective in terms of the overall execution time, and demonstrates the feasibility of automatic cluster assignment, processor set selection and data partition functions for data and task parallel programs.

The purpose of this paper is to propose novel parallel computational techniques for the parallelization of explicit finite element generalized approximate inverse methods, based on Portable Operating System Interface for UniX (POSIX) threads, for multicore systems.

The authors' main motive for the derivation of the new Parallel Generalized Approximate Inverse Finite Element Matrix algorithmic techniques is that they can be efficiently used in conjunction with explicit preconditioned conjugate gradient‐type schemes on multicore systems. The proposed parallelization technique of the Optimized Banded Generalized Approximate Inverse Finite Element Matrix (OBGAIFEM) algorithm is achieved based on the concept of the “fish bone” approach with the use of a thread pool pattern. Theoretical estimates on the computational complexity of the parallel generalized approximate inverse finite element matrix algorithmic techniques are also derived.

Application of the proposed method on a two‐dimensional boundary value problem is discussed and numerical results are given on a multicore system using POSIX threads. These results tend to become optimum and are favorably compared to corresponding results from multiprocessor systems, as presented in recent work by Gravvanis et al.

The proposed parallel explicit finite element generalized approximate inverse preconditioning, using approximate factorization and approximate inverse algorithms, is an efficient computational method that is valuable for computer scientists and for scientists and engineers in engineering computations.

The progress of parallel computing in Information Retrieval (IR) is reviewed. In particular we stress the importance of the motivation in using parallel computing for text retrieval. We analyse parallel IR systems using a classification defined by Rasmussen and describe some parallel IR systems. We give a description of the retrieval models used in parallel information processing. We describe areas of research which we believe are needed.

This paper presents a routing method for two dimensional transputer arrays particularly designed for parallel non‐linear and dynamic finite element analysis. Some general routing strategies with their strengths and weaknesses are discussed. The need for problem specific routing algorithms in transputer systems is considered and the communication requirements for parallel finite element analysis described. From the communication requirements the design parameters for the routing method are specified and the architectural design of the Router presented. An example of the use of the Router in the parallel non‐linear finite element analysis is given and the robustness of the routing methodology is illustrated by using arbitrary mappings of finite element subdomains distributed over a transputer array.

The purpose of this paper is to find a more performing and automated procedure for linking an identification algorithm implemented in a general‐purpose environment with a commercial finite‐element code for magnetic field analysis. In particular, the use of a multiprocessor computer makes it possible to perform parallel computations keeping the calculation time reasonably low.

The method is applied to identify the B‐H curve of anisotropic magnetic laminations in the direction normal to the sheet surface. In total, three different optimization methods have been applied. First an evolution strategy algorithm for solving the identification problem was used; then genetic algorithm (GA) was applied. The results obtained using different methods were compared and discussed. The computation time is reduced by adjusting the refinement of the FEM mesh.

The key point has been the use of a derivative‐free and global‐search oriented algorithm. Even if a starting point far from the solution is chosen, a suitably large initial value of the search radius makes the convergence possible. The effect of the historical parameter of the minimization algorithm on convergence has also been investigated.

The main new idea presented in this paper is equipping a GA‐based identification procedure with an additional objective function describing the sensitivity of the flux density against a small perturbation in parameters. This approach gives a multiple objective problem which introduces possibility of choosing a compromise solution among many optimal solutions instead of only one, as in classical GA optimization algorithm. The paper is mainly addressed to readers interested in the efficient use of GA‐based identification.

Components and functions which comprise a computer, such as address registers and cycle time, are defined and described. Characteristics of 8‐, 16‐, and 32‐bit machines and the confusing use of multiple bit‐size architecture in some computers are detailed. Appropriate applications for each bit‐size are suggested. Detailed explanations are provided to amplify statements in the text. This article should be read twice for maximum comprehension, once reading the text and explanations together, and once reading only the text.