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.

This paper gives a bibliographical review of the finite element and boundary element parallel processing techniques from the theoretical and application points of view. Topics include: theory – domain decomposition/partitioning, load balancing, parallel solvers/algorithms, parallel mesh generation, adaptive methods, and visualization/graphics; applications – structural mechanics problems, dynamic problems, material/geometrical non‐linear problems, contact problems, fracture mechanics, field problems, coupled problems, sensitivity and optimization, and other problems; hardware and software environments – hardware environments, programming techniques, and software development and presentations. The bibliography at the end of this paper contains 850 references to papers, conference proceedings and theses/dissertations dealing with presented subjects that were published between 1996 and 2002.

A fully parallel algorithm for the solution of a finite element system using a MIMD (multiple‐instruction multiple‐data architecture) parallel computer is presented. The formulation includes a simple domain decomposer that automatically divides a finite element mesh into a list of subdomains to guarantee the load balancing. Furthermore, each subdomain is assigned to a processor of a parallel computer and treated as a sub‐finite element system with information exchanged through the interface between two adjacent subdomains. With this new algorithm, these sub‐finite element systems are solved fully parallelly as independent finite element systems, not only the computations of the interior nodes but also the computations of the interface nodes can be executed parallelly. Also, the inherently sequential Gauss‐Seidel and SOR schemes are altered into fully parallel iterative schemes. An implementation of this new scheme on an iPSC/2 D5 Hypercube Concurrent Computer reached an efficiency of more than 100% when compared with the sequential SOR scheme.

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.

The purpose of this paper is to propose general parallelism techniques for holistic twig join algorithms to process queries against Extensible Markup Language (XML) databases on a multi‐core system.

The parallelism techniques comprised data and task parallelism. As for data parallelism, the paper adopted the stream‐based partitioning for XML to partition XML data as the basis of parallelism on multiple CPU cores. The XML data partitioning was performed in two levels. The first level was to create buckets for creating data independence and balancing loads among CPU cores; each bucket was assigned onto a CPU core. Within each bucket, the second level of XML data partitioning was performed to create finer partitions for providing finer parallelism. Each CPU core performed the holistic twig join algorithm on each finer partition of its own in parallel with other CPU cores. In task parallelism, the holistic twig join algorithm was decomposed into two main tasks, which were pipelined to create parallelism. The first task adopted the data parallelism technique and their outputs were transferred to the second task periodically. Since data transfers incurred overheads, the size of each data transfer needed to be estimated cautiously for achieving optimal performance.

The data and task parallelism techniques contribute to good performance especially for queries having complex structures and/or higher values of query selectivity. The performance of data parallelism can be further improved by task parallelism. Significant performance improvement is attained by queries having higher selectivity because more outputs computed by the second task is performed in parallel with the first task.

The proposed parallelism techniques primarily deals with executing a single long‐running query for intra‐query parallelism, partitioning XML data on‐the‐fly, and allocating partitions on CPU cores statically. During the parallel execution, presumably there are no such dynamic XML data updates.

The effectiveness of the proposed parallel holistic twig joins relies fundamentally on some system parameter values that can be obtained from a benchmark of the system platform.

The paper proposes novel techniques to increase parallelism by combining techniques of data and task parallelism for achieving high performance. To the best of the author's knowledge, this is the first paper of parallelizing the holistic twig join algorithms on a multi‐core system.

In this paper, a parallel preconditioning technique based on the additive variant of overlapping domain decomposition is described and implemented to solve magnetostatic field problems. This technique involves covering the domain with a number of overlapping subdomains. The pre‐conditioner results from adding together approximate inversions on the subdomains, Theoretical estimates for the rate of convergence for the resulting algorithm are available and are based on the properties of underlying differential equations. Numerical experiments are given to demonstrate the effectiveness of this algorithm.

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.

This paper presents a multi‐level frontal algorithm and its implementation and applications on parallel computation. A multi‐frontal program is given which may be used for unsymmetric finite element matrix equations. The parallel program is developed on a cluster of workstations. The PVM (parallel virtual machine) system is used to handle communications among networked workstations. The method has advantages such as numbering of the finite element mesh in an arbitrary manner, simple programming organisation, smaller core requirements and computation times. An implementation of this parallel method on workstations is discussed, the speedup and efficiency of this method being demonstrated and compared with general domain decomposition method based on band matrix methods by numerical examples.

This paper aims to propose a parallel-in-space-time finite-element method (FEM) for transient motor starting analyses. Although the domain decomposition method (DDM) is suitable for solving large-scale problems and the parallel-in-time (PinT) integration method such as Parareal and time domain parallel FEM (TDPFEM) is effective for problems with a large number of time steps, their parallel performances get saturated as the number of processes increases. To overcome the difficulty, the hybrid approach in which both the DDM and PinT integration methods are used is investigated in a highly parallel computing environment.

First, the parallel performances of the DDM, Parareal and TDPFEM were compared because the scalability of these methods in highly parallel computation has not been deeply discussed. Then, the combination of the DDM and Parareal was investigated as a parallel-in-space-time FEM. The effectiveness of the developed method was demonstrated in transient starting analyses of induction motors.

The combination of Parareal with the DDM can improve the parallel performance in the case where the parallel performance of the DDM, TDPFEM or Parareal is saturated in highly parallel computation. In the case where the number of unknowns is large and the number of available processes is limited, the use of DDM is the most effective from the standpoint of computational cost.

This paper newly develops the parallel-in-space-time FEM and demonstrates its effectiveness in nonlinear magnetoquasistatic field analyses of electric machines. This finding is significantly important because a new direction of parallel computing techniques and great potential for its further development are clarified.

The numerical simulation of two‐dimensional incompressible complex flows of viscoelastic fluids is presented. The context is one, relevant to the food industry (dough kneading), of stirring within a cylindrical vessel, where stirrers are attached to the lid of the vessel. The motion is driven by the rotation of the outer vessel wall, with various stirrer locations. With a single stirrer, both a concentric and an eccentric configuration are considered. A double‐stirrer eccentric case, with two symmetrically arranged stirrers, is also contrasted against the above. A parallel numerical method is adopted, based on a finite element semi‐implicit time‐stepping Taylor‐Galerkin/pressure‐correction scheme. For viscoelastic fluids, constant viscosity Oldroyd‐B and two shear‐thinning Phan‐Thien/Tanner constitutive models are employed. Both linear and exponential models at two different material parameters are considered. This permits a comparison of various stress, shear and extensional properties and their respective influences upon the flow fields generated. Variation with increasing speed of vessel and change in mixer geometry are analysed with respect to the flow kinematics and stress fields produced. Optimal kneading scenarios are commended with asymmetrical stirrer positioning, one‐stirrer proving better than two. Then, models with enhanced strain‐hardening, amplify levels of localised maxima in rate‐of‐work done per unit power consumed. Simulations are conducted via distributed parallel processing, performed on work‐station clusters, employing a conventional message passing protocol (PVM). Parallel results are compared against those obtained on a single processor (sequential computation). Ideal linear speed‐up with the number of processors has been observed.